DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
This Office action is in response to correspondence received June 25, 2026.
Claims 1 and 11 are amended. Claims 1-20 are pending and have been examined.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim(s) recite(s):
Claim 1:
receive a plurality of alerts relating to a building, the plurality of alerts comprising alert types; identify a set of alert disposition options for the plurality of alerts based on the alert types; estimate probabilities of use for the set of alert disposition options based on selected alert disposition options for a plurality of historical alerts related to the building; calculate, for the set of alert disposition options, alert disposition risk scores using the estimated probabilities of use of the set of alert disposition options; calculate, for the plurality of alerts, alert risk scores based on a combination of the alert disposition risk scores for the set of alert disposition options of the plurality of alerts; present two or more of the plurality of alerts based on the alert risk scores; automatically execute, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of the building equipment of the building associated with the alert.
Claim 11:
A method of operating a facility security system, the method comprising: receiving a plurality of alerts, wherein a first alert of the plurality of alerts comprises an alert activation signal, an alert type, and a set of alert contextual data; identifying a set of alert disposition options for the first alert based on the alert type; classifying an alert disposition option for the first alert estimates a probability of use of the alert disposition option within the set of alert disposition options initially trained using historical alert data, determining an alert risk score for the first alert, wherein the alert risk score aggregates one or more risk model outputs, further wherein the one or more risk model outputs is based on an alert disposition option classification, a level of security interest of the alert disposition option, and a cost of loss of an asset monitored by the facility security system; prioritizing the first alert based on the alert risk score; presenting, a prioritized list of the plurality of alerts, the prioritized list comprising the plurality of alerts, alert risk scores, and alert disposition options; recording the alert disposition option selected by a user for the first alert; and storing the recorded alert disposition option and estimate probability of use for the set of alert disposition options based on the alert disposition option selected by the user for the first alert; automatically executing, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of the building equipment of the building associated with the alert.
This judicial exception is not integrated into a practical application. The additional elements taken in combination amount instructions to apply generic computer elements such as memory, processor, and user interface to the abstract idea. See MPEP 2106.05(f)(2). Taken together they amount to a generic computer (bc all generic computers – mobile phones, laptops etc – have them); from a plurality of sensors disposed within a building, wherein the sensors are described in par 59 as “include smoke detectors, temperature and humidity sensors, door position sensors, window position sensors, occupancy detectors, etc. In some embodiments, sensors 332 may also include sensors that are internal to building equipment, such as speed and temperature sensors for a chiller of an HVAC system”; automatically executing an action; as well as a classifier engine where an engine is an input-output element that, like a function, takes an input and produces an output; and transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state. Additionally the classifier model is trained which merely describes which kind of classifier engine is being used (a trained one), and per claim 11 the classifier engine is retrained however as described in Example 47 claim 2 of the Subject Matter Eligibility training examples, see ttps://www.uspto.gov/patents/laws/examination-policy/subject-matter-eligibility , which was found to be ineligible because the training at its broadest reasonable interpretation is a mathematical calculation. Alternatively it is an apply it limitation because the retraining describes generally the kind of data used but there is no detail as to how the retraining is performed (see MPEP 2106.05(f)(1) – a desired outcome or result is what is claimed). The initially trained (claim 11) and trained (claim 1) are not positively recited steps as under a broadest reasonable interpretation they describe something that occurred in the past (an out of the box solution that is applied in the claims).
Taken together these amount to no more than a combination of elements operating in their ordinary capacity. Automatic execution is performing a programmed step, such as using a generic computer. Changing an operational state of building equipment which is described as any other type of system or device see par 059 can include displaying information on a computer which is a device. In combination with the classifier engine which is software only described by its functional result see MPEP 2106.05(f)(1) is further software added to the generic computing system. Taken in combination this amounts to no more than using a computer to automate mental process steps. Taken in combination and the claims as a whole this amounts to a mental process of decision making based on information inputs that is applied to information from sensors, analyzed by applied classifier engine that has no technical detail as to how it works, and then a result is output where a task is automatically executed, which could be a computer talking to another computer. Even if the task that was automatically executed was an access control system, surveillance system, external or remote security system, etc as described in par 59, it is not clear that this would overcome the 101 rejection as there is no technical detail or further limiting of technical detail apart from sending information from one system or similar nonce-type term to another system or nonce-type term. Therefore, for these reasons, the abstract idea is not integrated into a practical application.
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the reasoning in the practical application section is applied here: for the same reason that the combination of elements is not a practical application, the combination is not significantly more than the abstract idea.
Per claims 2 and 12, which are similar in scope, the abstract idea is further defined with an alert severity rating, an information element that one could choose mentally to use to describe a number or condition.
Per claims 3 and 13, like 2 and 12, these are labels for information that one could decide to use to describe alerts.
Per claims 4 and 14, which are similar in scope, using a table to choose actions is like using a paper spreadsheet as that would teach table, and that it is stored on a system teaches using a generic computer to store information, like a PC storing an excel file.
Per claims 5 and 15, which are similar in scope, choosing options that are a assigned a code is a further mental process step.
Per claims 6 and 16, determining scores based on different inputs is a mental process step as it consists of preselected rules to make a final decision, like an equation. This could be done mentally or with pen and paper as they could simply be binary choices that amount to a score, or a linear equation, or a combination of both, which are readily done mentally or with pen and paper. That an engine is used is instructions to apply the abstract idea (here the linear equation) to a computer (engine is shorthand for a function).
Per claims 7 and 17, which are similar in scope, the limitation that a calculation or result is arrived at “by” a machine learning model that comprises one or more of named algorithms is an apply it limitation, instructing the user to apply the ML models to the abstract idea. This is because there is no detail as to how these models would arrive at the probability, only that they used in some way to do so.
Per claims 8 and 18, which are similar in scope, the abstract idea is further defined by the alert data comprising different sources, which could be observed. Using ML models named solely for their desired functional output (there is no indication what a “door classification model” would look like algorithmically except that the output would classify doors) is further apply it limitations.
Per claims 9 and 19, which are similar in scope, the additional element of the classification engine being, in the past (not positively claimed), trained on historical data is an apply it limitation because training using historical (alert) data is essentially describing training as it is usually done on previous data. Therefore this is a further apply it limitation.
Per claims 10 and 20, which are similar in scope, the limitation is similar to 9 and 19 where the engine is retrained using “contemporary alert data automatically collected by the system.” The retraining here is similar to the training and using the trained model as described in Example 47 claim 2 of the Subject Matter Eligibility training examples, see https://www.uspto.gov/patents/laws/examination-policy/subject-matter-eligibility , which was found to be ineligible because the training at its broadest reasonable interpretation is a mathematical calculation. Alternatively it is an apply it limitation because the retraining describes generally the kind of data used but there is no detail as to how the retraining is performed (see MPEP 2106.05(f)(1) – a desired outcome or result is what is claimed).
Therefore, claims 1-20 are rejected under 35 USC 101.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-5, 9, and 10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Fuller et al., US Pat. No. 9754478 B1 (“Fuller”) in view of Beale et al., US Pat No 10762773 B1 ("Beale"), further in view of Trundle US Pat No 9013294 B1 ("Trundle").
Per claim 1, Fuller teaches A building security system comprising: one or more memory devices configured to store instructions that, when executed by one or more processors, cause the one or more processors to: receive, from a plurality of sensors disposed within a building, a plurality of alerts relating to the building in col 2 ln 60 - col 3 ln 26: "Computing device 106 can receive a notification of an alarm from building automation system 108. For example, computing device 106 can receive an alarm relating to a fault that may be occurring in building automation system 108. For instance, a fault in a piece of HVAC equipment such as a stuck valve may be occurring that causes an alarm to be generated and sent to building automation system 108 by a sensor that sensed the stuck valve. After receiving the alarm from building automation system 108, computing device 106 can transmit a notification of the alarm to the number of mobile devices 112-1, 112-2, and/or 112-N via network 110. As used herein, a mobile device can be a phone (e.g., a smart phone), a tablet, a personal digital assistant (PDA), and/or a wearable device such as a wrist-worn device (e.g., a smartwatch) and/or a head-worn device (e.g., smart-glasses), among other types of devices that may be carried and/or worn by a user. As used herein, a fault can include an event that occurs to cause a piece of equipment and/or a control strategy of a building to function improperly or to cause abnormal behavior in a building, or a zone of the building, serviced by building automation system 108. In some examples, a fault can include a piece of equipment breaking down. In some examples, a fault can include a component of a piece of equipment ceasing to function correctly. In some examples, a fault can include abnormal behavior of a piece of equipment and/or a zone. Although a fault is described as including equipment breakdowns and abnormal behavior, embodiments of the present disclosure are not so limited. For example, faults can include any other event that causes equipment or control strategies to function improperly, and/or causes abnormal behavior to occur in a building serviced by building automation system 108." See also col 2 ln 26-50. Plurality taught in Fig 2 as well, see four different alerts from different sensors.
Fuller then teaches the plurality of alerts comprising alert types in col 4 ln 30-56: "Each of the alarms generated by building automation system 108 can includes a number of attributes. As used herein, attributes of an alarm can include characteristics of an alarm. Attributes of an alarm can be assigned to the alarm by the building automation system 108. Attributes of an alarm can be used to determine whether an additional alarm from building automation system 108 is similar to a suppressed alarm and should also be suppressed, as will be further described herein. Attributes of an alarm can include a condition of the alarm, such as the issue state of the alarm. An issue state of an alarm can include a parameter to describe what the issue state is. For instance, an issue state can include FAULT (e.g., a fault), DISCONNECTED (e.g., equipment is disconnected), POWER OFF (e.g., equipment is powered off), OFFLINE (e.g., equipment is offline), OFF-LIMIT (e.g., equipment parameter/variable off limit), PV-HIGH (e.g., present value of parameter/variable too high), etc. In some examples, a condition of an alarm can include an alarm being a new alarm. For instance, the alarm may be regarding a fault in a piece of equipment included in a building that has not generated an alarm before, or has not generated an alarm for a specified period of time (e.g., has not generated an alarm for one week or longer). In some examples, a condition of an alarm can include the alarm having recently occurred and is therefore being repeated. In some examples, a condition of an alarm can include the alarm expiring." See also col 4 ln 57 - col 5 ln 10. Issue states teaches alert types.
Fuller then teaches identify a set of alert disposition options for the plurality of alerts based on the alert types in col 4 ln 57 - col 5 ln 2: "Attributes of an alarm can include a priority of the alarm. In some examples, a priority of an alarm can include an urgent priority. For instance, the alarm may be regarding a fault in a piece of critical equipment included in a building, such as equipment included in a security system or HVAC system. In some examples, a priority of an alarm can include a high priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be important, but may not need to be addressed in an immediate fashion. In some examples, a priority of an alarm can include a low priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be ignored for a longer period of time than a fault corresponding to a high or urgent priority alarm." Urgent teaches one disposition, to be fixed immediately. High another disposition, not needed to be addressed immediately. Low another.
Fuller then teaches present two or more of the plurality of alerts based on [information] in col 8 ln 60-67: " As shown in FIG. 2, alarm attributes 218 can include conditions, priority, source, and/or category of the alarm. For example, the notification of an alarm 216 can include alarm attributes 218 that specify a condition of the alarm (e.g., 19.90 M3/HR flow rate) and a source of the alarm (e.g., PASS 1 FLOW TO HEATER). This information can allow a user to determine actions to take regarding the alarm and/or how quickly to take those actions." Plurality of alerts are presented in Fig 2.
Fuller does not teach estimate probabilities of use for the set of alert disposition options using a classifier model trained based on selected alert disposition options for a plurality of historical alerts related to the building; calculate, for the set of alert disposition options, alert disposition risk scores using the estimated probabilities of use of the set of alert disposition options; calculate, for the plurality of alerts, alert risk scores based on a combination of the alert disposition risk scores for the set of alert disposition options of the plurality of alerts present an alert based on the alert risk score.
Beale teaches a false alarm predicting model. See abstract.
Beale teaches estimate probabilities of use for the set of alert disposition options using a classifier model trained based on selected alert disposition options for a plurality of historical alerts related to the building in col 5 ln 64 - col 6 ln 12: "In some embodiments, the learning module can identify a score to determine whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm. For example, the score can be indicative of a likelihood or a probability that the combination represents the false alarm or the valid alarm. In some embodiments, the score can be based on an amount by which the alarm signal and the additional information match the plurality of alarm signals from the historical time period and the plurality of additional information from the historical time period, and in some embodiments, the alarm signal and/or the additional information can be automatically or manually assigned different weights for such a matching comparison." Assigning weights to perform a matching comparison teaches a classifier model because the model is determining if an alarm is false or valid (the classification).
Beale then teaches calculate, for the set of alert disposition options, alert disposition risk scores using the estimated probabilities of use of the set of alert disposition options in col 5 ln 64 - col 6 ln 12: "In some embodiments, the learning module can identify a score to determine whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm. For example, the score can be indicative of a likelihood or a probability that the combination represents the false alarm or the valid alarm. In some embodiments, the score can be based on an amount by which the alarm signal and the additional information match the plurality of alarm signals from the historical time period and the plurality of additional information from the historical time period, and in some embodiments, the alarm signal and/or the additional information can be automatically or manually assigned different weights for such a matching comparison." This teaches risk score.
Beale then teaches calculate, for the plurality of alerts, alert risk scores based on a combination of the alert disposition risk scores for the set of alert disposition options of the plurality of alerts in col 6 ln 27-39: " However, in some embodiments, the score can include a range of values with a calculated distribution (e.g. Gaussian) that indicates whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm. In such embodiments, the automated dispatcher module can include a cumulative distribution function that indicates when the automated dispatcher module should alert the user and/or the authorities, and in some embodiments, a sensitivity of the automated dispatcher module to the score can be automatically or manually adjusted based on the user preference data, such as days of the week or when the user is out of town.” This teaches the equivalent of calculate, for the plurality of alerts, alert risk scores based on a combination of the alert disposition risk scores for the set of alert disposition options of the plurality of alerts as it is a distribution function wherein the score includes a range of values (“combination of the alert disposition risk scores for the set of alert disposition options of the plurality of alerts”)
Beale then teaches present an alert based on the alert risk score col 6 ln 14-22: " Then, the automated dispatcher module can compare the score to a threshold value to automatically determine whether to alert the user and/or the relevant authorities about the alarm signal. When such a comparison and/or the score indicates that the automated dispatcher module should alert the user and/or the relevant authorities, the automated dispatcher module can automatically alert the user and/or the relevant authorities about the alarm signal without human intervention."
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the alert risk score teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Fuller does not teach: and automatically execute, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of building equipment of the building associated with a sensor of the plurality of sensors corresponding to the alert, the follow-up action comprising transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state.
Trundle teaches an alarm probability score and handling an event based on the score.
Trundle teaches and automatically execute, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of building equipment of the building associated with a sensor of the plurality of sensors corresponding to the alert, the follow-up action comprising transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state in col 9 ln 26-51: "In the example shown in FIG. 1C, the control panel 20 detects a basement door opening event at 2:00 AM on a Monday based on receiving an indication from the basement door sensor 22. The control panel 20 provides an alert to the monitoring server 30 indicating the detected alarm event, and the monitoring server 30 proceeds to access data relevant to the detected alarm event. The monitoring server accesses data that includes data indicating (1) that there was no interior motion detected by the motion sensor 24, (2) that the alarm system monitoring the property 10 detected the phone line being cut prior to the basement door sensor 22 detecting the basement door being opened, (3) that historical sensor data collected from the alarm system indicates that the basement door is never opened around 2:00 AM, and (4) that there has been a recent spike in night time break-ins for the region of the property 10. The monitoring system 30 includes these factors in the alarm probability factors 40, where the monitoring system 30 then evaluates the factors to determine an alarm probability score estimating the likelihood that basement door opening at 2:00 AM on Monday should be treated as an emergency situation. Based on the evaluation, the monitoring server 30 determines an alarm probability score of 95% associated with the detected alarm event, and subsequently determines how to handle the detected alarm event based on the determined alarm probability score."col 11 ln 43-42: "The monitoring system control unit 110 communicates with the module 122 and the camera 130 to perform surveillance or monitoring. The module 122 is connected to one or more lighting systems and is configured to control operation of the one or more lighting systems. The module 122 may control the one or more lighting systems based on commands received from the monitoring system control unit 110. For instance, the module 122 may cause a lighting system to illuminate an area to provide a better image of the area when captured by a camera 130." Two or more is taught by basement door being opened and phone line being cut. Follow up action change in operational state taught by lights being turned on in an area of the property.
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify to modify the alert teachings of Fuller with the follow up action based on two alerts teaching of Trundle because Trundle teaches in col 1 ln 57-67: “False alarms are a significant issue for security systems. Security systems often detect events that reflect potential alarm conditions, but that are not in fact alarm situations. When a false alarm is detected by a security system and no one is available to confirm that the alarm is false, emergency services may be dispatched unnecessarily. Dispatching emergency services for false alarms consumes resources of emergency personnel and limits the ability of emergency personnel to respond to real alarm situations.” As Trundle’s teaching would save resources, one would be motivated to modify Fuller with Trundle so fewer resources would be consumed.
Per claim 2, Fuller, Beale, and Trundle teach the limitations of claim 1, above. Fuller further teaches wherein the alert types are associated with alert severity ratings in col 4 ln 57 – col 5 ln 2: “Attributes of an alarm can include a priority of the alarm. In some examples, a priority of an alarm can include an urgent priority. For instance, the alarm may be regarding a fault in a piece of critical equipment included in a building, such as equipment included in a security system or HVAC system. In some examples, a priority of an alarm can include a high priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be important, but may not need to be addressed in an immediate fashion. In some examples, a priority of an alarm can include a low priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be ignored for a longer period of time than a fault corresponding to a high or urgent priority alarm”
Per claim 3, Fuller, Beale, and Trundle teach the limitations of claim 1, above. Fuller further teaches wherein the plurality of alerts further comprise a set of alert contextual data, the set of alert contextual data comprising a set of alert metadata, a set of threat data, and a set of facility data in col 5 ln 2-34: “Attributes of an alarm can include a source of the alarm. In some examples, the source of an alarm may include the source of the equipment with a fault. For instance, an alarm may be generated as a result of a fault in a piece of HVAC equipment, and the source of the fault may be a valve of a cooling coil of an air handling unit (AHU) is stuck open, a radiator valve is stuck shut, a hot water pump has stopped working, etc. In some examples, the source of an alarm may include the location of the equipment generating the fault. For instance, an alarm may be generated as a result of a fault in a piece of HVAC equipment, and the source of the fault may be the cooling coil of the AHU located on a third floor equipment room of a building, a radiator valve located in a first floor classroom of a building, or a hot water pump located in an underground boiler room of the building, etc.
Attributes of an alarm can include a category of the alarm. In some examples, the category of an alarm may include the category of the equipment with a fault. For instance, an alarm may be generated as a result of a fault in a piece of HVAC equipment, a fault in a security system, a fault in an electrical system, a fault in a plumbing system, etc. In some examples, the category of an alarm may include a Point Alarm (e.g., an alarm regarding a part of a piece of equipment), System Alarm (e.g., software or core issues such as an expired software license, etc.), and/or other categories of alarms.
Although attributes of the alarms are described as including a condition, priority, source, and category of the alarms, embodiments of the present disclosure are not so limited. For example, attributes of the alarms can include other alarm descriptors/characteristics.”
Fuller does not teach a set of environmental data .
Beale teaches a set of environmental data in col 3 ln 21-28: “In accordance with disclosed embodiments, the security system can protect a geographic area, and in some embodiments, the additional information can include weather data from a time associated with the alarm signal, movement data associated with the geographic area during the time associated with the alarm signal, a location of users of the security system during the time associated with the alarm signal, and/or incident reports relevant to the geographic area.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the set of environmental data teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Per claim 4, Fuller, Beale, and Trundle teach the limitations of claim 1, above. Fuller further teaches wherein the set of alert disposition options for the alert types comprises a table of one or more actions that a user may select to dispose of an alert, wherein the table is stored on the one or more memory devices of the building security system in col 6 ln 6-20: “In some examples, a user may have previously suppressed notifications of an alarm regarding a fault of a valve included in HVAC equipment; in response to computing device 106 determining that the attributes of another alarm regarding the fault of the valve received from building automation system 108 match those of the previously suppressed alarm regarding the fault of the valve, computing device 106 can refrain from transmitting a notification of the alarm regarding the fault of the valve to a mobile device among the number of mobile devices 112-1, 112-2, 112-N.” The user, selecting to suppressed, is a table of an action that a user may select as the action is in a database, which is a table.
Per claim 5, Fuller, Beale, and Trundle teach the limitations of claim 4, above. Fuller further teaches wherein the options in the set of alert disposition options are assigned a code, the code indicating a level of security significance in col 9 ln 1 – 16: “After a user has received the notification of alarm 216 and has considered actions to take regarding the alarm, the user can suppress further notifications for the alarm and/or notifications of other alarms with similar attributes to the alarm using suppress options 220. For example, a user can utilize suppress options 220 to suppress further notifications for the alarm and/or notifications of other alarms with similar attributes to the alarm for a predetermined length of time, as will be further described in connection with FIG. 3.
Once a user has elected to suppress notifications for the alarm and/or notifications of other alarms with similar attributes to the alarm using suppress options 220, a computing device (e.g., computing device 106, previously described in connection with FIG. 1) can receive the instructions to suppress notifications of the alarm and/or notifications of other alarms with similar attributes to the alarm.” The suppress options under a broadest reasonable interpretations teach a code because it can either suppress notification for the same alarm or for that alarm and similar alarms.
Per claim 9, Fuller, Beale, and Trundle teach the limitations of claim 1, above. Fuller does not teach wherein the classifier model is trained to calculate the probability of use of an alert disposition option within the set of alert disposition options using historical alert data
Beale teaches wherein the classifier model is trained to calculate the probability of use of an alert disposition option within the set of alert disposition options using historical alert data In col 6 ln 2-22: “For example, the score can be indicative of a likelihood or a probability that the combination represents the false alarm or the valid alarm. In some embodiments, the score can be based on an amount by which the alarm signal and the additional information match the plurality of alarm signals from the historical time period and the plurality of additional information from the historical time period, and in some embodiments, the alarm signal and/or the additional information can be automatically or manually assigned different weights for such a matching comparison. Furthermore, the learning module can transmit the score to the automated dispatcher module, for example, with the status signal. Then, the automated dispatcher module can compare the score to a threshold value to automatically determine whether to alert the user and/or the relevant authorities about the alarm signal. When such a comparison and/or the score indicates that the automated dispatcher module should alert the user and/or the relevant authorities, the automated dispatcher module can automatically alert the user and/or the relevant authorities about the alarm signal without human intervention.” See also col 5 ln 20-29: “In some embodiments, any of the feedback signals described herein can include user input explicitly identifying the alarm signal or the plurality of alarm signals from the historical time period as the valid alarm or the false alarm. Additionally or alternatively, in some embodiments, any of the feedback signals described herein can include information related to actions executed in response to the alarm signal or the plurality of alarm signals from the historical time period that are indicative of the valid alarm or the false alarm.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the alert risk score teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Per claim 10, Fuller, Beale, and Trundle teach the limitations of claim 1, above. Fuller does not teach wherein the classifier model is periodically retrained to calculate the probability of use of an alert disposition option within the set of alert disposition options using contemporary alert data automatically collected by the system.
Beale teaches the classifier model is periodically retrained to calculate the probability of use of an alert disposition option within the set of alert disposition options using contemporary alert data automatically collected by the system in col 5 ln 12-17: “For example, in some embodiments, the learning module can receive feedback signals indicating whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm and can use those feedback signals to update the false alarm predicting model for the increased accuracy at the future times.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the alert risk score teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Claim(s) 6, 8, 11-15, 19, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Fuller et al., US Pat. No. 9754478 B1 (“Fuller”) in view of Beale et al., US Pat No 10762773 B1 ("Beale"), further in view of Trundle US Pat No 9013294 B1 ("Trundle"), further in view of Pourmohammad et al., US PGPUB 20190138512 A1 (“Pourmohammad”).
Per claim 6, Fuller, Beale, and Trundle teach the limitations of claim 1, above. Fuller does not teach wherein the alert risk scores are determined by a dynamic prioritization model based on inputs comprising an alert type, alert contextual data, a level of security interest, a cost of an asset, a disposition probability, and alert disposition codes applied by a user.
Beale teaches wherein the alert risk scores are determined by a dynamic prioritization model based on inputs comprising an alert type, in col 4 ln 42-45” “For example, the other customized parameters can include a defined geographic area, a type of the plurality of alarm signals”
alert contextual data, in col 3 ln 65 – col 4 ln 5: “With the global model, in some embodiments, the plurality of additional information from the historical time period can include the weather data from the time associated with one of the plurality of alarm signals from the historical time period,”
a disposition probability, in col 5 ln 12-19: “For example, in some embodiments, the learning module can receive feedback signals indicating whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm and can use those feedback signals to update the false alarm predicting model for the increased accuracy at the future times.”
a level of security interest, in col 7 ln 16-20: “Additionally or alternatively, in some embodiments, a sensitivity of the central monitoring station to the score can be automatically or manually adjusted based on a price or level of service that the central monitoring station provides to the user.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the set of codes and alerts teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Trundle teaches and alert disposition codes applied by a user in col 16 ln 16-27: “In these implementations, the camera 130 may be set to capture images on a periodic basis when the alarm system is armed in an "Away" state, but set not to capture images when the alarm system is armed in a "Stay" state or disarmed. In addition, the camera 130 may be triggered to begin capturing images when the alarm system detects an event, such as an alarm event, a door opening event for a door that leads to an area within a field of view of the camera 130, or motion in the area within the field of view of the camera 130. In other implementations, the camera 130 may capture images continuously, but the captured images may be stored or transmitted over a network when needed.”
Pourmohammad teaches a building management system with a dynamic risk score. See abstract.
Pourmohammad teaches a cost of an asset in par 0323: “Regardless of the imminent threat and its nature, the value of the asset is important in evaluating the risk to the owner. The asset value becomes more important when a company has multiple types of assets with different functionality and responsibilities. Some of them might be strategic and very valuable. But, others might be smaller and less valuable compared to the others. Asset assessment includes the asset cost estimation besides vulnerability assessment. The result of the asset value assessment is translated to a number between 1 to 10 in the risk model to represent the least to most valuable assets.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the cost of an asset in determining risk score teaching of Pourmohammad because Pourmohammad teaches in par 0211 that: “The analytics systems and methods as described herein can generate risk information for use in prioritization of alarms, presenting users with contextual threat and/or asset information, reducing the response time to threats by raising the situational awareness, and automating response actions.” As it would present useful information to users and reduce time to threats and automate response actions, one would be motivated to modify Fuller with Pourmohammad.
Per claim 8 , Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claim 6, above. Fuller does not teach wherein the alert contextual data comprise internal contextual data and outputs of one or more machine learning models, the one or more machine learning models further comprising a spatial model, an occupancy model, a door classification model, and a sensor state model.
Pourmohammad teaches wherein the alert contextual data comprise internal contextual data and outputs of one or more machine learning models, the one or more machine learning models further comprising a spatial model, an occupancy model, a door classification model, and a sensor state model in par 0249: “The result of the enrichment by the asset information enricher 302 is the enriched threat 308. The enriched threat 308 can include an indication of a threat, an indication of an asset affected by the threat, and contextual information of the asset and/or threat. The RAP 120 includes risk engine 310 and risk score enricher 312. Risk engine 310 can be configured to generate a risk score (or scores) for the enriched threat 308. Risk engine 310 can be configured to generate a dynamic risk score for the enriched threat 308. The risk score enricher 312 can cause the dynamic risk can be included in the enriched threat 316 generated based on the enriched threat 308.” See also par 0317: “The risk engine 310 of the RAP 120 can be configured to generate risk scores for the threats via a model. The model used by the risk engine 310 can be based on Expected Utility Theory and formulated as an extended version of a Threat, Vulnerability and Cost (TVC) model. The risk engine 310 can be configured to determine the risk scores on a per asset basis. The threats can all be decoupled per asset in the processing pipeline as well as the calculation of the risk. For example, if a protest or weather condition is created alerts towards multiple buildings, separate alerts per building will be generated based on the geo-fences of the building and the detected alert.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the cost of an asset in determining risk score teaching of Pourmohammad because Pourmohammad teaches in par 0211 that: “The analytics systems and methods as described herein can generate risk information for use in prioritization of alarms, presenting users with contextual threat and/or asset information, reducing the response time to threats by raising the situational awareness, and automating response actions.” As it would present useful information to users and reduce time to threats and automate response actions, one would be motivated to modify Fuller with Pourmohammad.
Per claim 11, Fuller teaches A method of operating a facility security system, the method comprising: receiving, from a plurality of sensors, a plurality of alerts, in col 2 ln 60 - col 3 ln 26: "Computing device 106 can receive a notification of an alarm from building automation system 108. For example, computing device 106 can receive an alarm relating to a fault that may be occurring in building automation system 108. For instance, a fault in a piece of HVAC equipment such as a stuck valve may be occurring that causes an alarm to be generated and sent to building automation system 108 by a sensor that sensed the stuck valve. After receiving the alarm from building automation system 108, computing device 106 can transmit a notification of the alarm to the number of mobile devices 112-1, 112-2, and/or 112-N via network 110. As used herein, a mobile device can be a phone (e.g., a smart phone), a tablet, a personal digital assistant (PDA), and/or a wearable device such as a wrist-worn device (e.g., a smartwatch) and/or a head-worn device (e.g., smart-glasses), among other types of devices that may be carried and/or worn by a user. As used herein, a fault can include an event that occurs to cause a piece of equipment and/or a control strategy of a building to function improperly or to cause abnormal behavior in a building, or a zone of the building, serviced by building automation system 108. In some examples, a fault can include a piece of equipment breaking down. In some examples, a fault can include a component of a piece of equipment ceasing to function correctly. In some examples, a fault can include abnormal behavior of a piece of equipment and/or a zone. Although a fault is described as including equipment breakdowns and abnormal behavior, embodiments of the present disclosure are not so limited. For example, faults can include any other event that causes equipment or control strategies to function improperly, and/or causes abnormal behavior to occur in a building serviced by building automation system 108." See also col 2 ln 26-50. Plurality taught in Fig 2 as well, see four different alerts from different sensors.
Fuller then teaches wherein a first alert of the plurality of alerts comprises an alert activation signal, an alert type, and a set of alert contextual data in col 4 ln 30-56: "Each of the alarms generated by building automation system 108 can includes a number of attributes. As used herein, attributes of an alarm can include characteristics of an alarm. Attributes of an alarm can be assigned to the alarm by the building automation system 108. Attributes of an alarm can be used to determine whether an additional alarm from building automation system 108 is similar to a suppressed alarm and should also be suppressed, as will be further described herein. Attributes of an alarm can include a condition of the alarm, such as the issue state of the alarm. An issue state of an alarm can include a parameter to describe what the issue state is. For instance, an issue state can include FAULT (e.g., a fault), DISCONNECTED (e.g., equipment is disconnected), POWER OFF (e.g., equipment is powered off), OFFLINE (e.g., equipment is offline), OFF-LIMIT (e.g., equipment parameter/variable off limit), PV-HIGH (e.g., present value of parameter/variable too high), etc. In some examples, a condition of an alarm can include an alarm being a new alarm. For instance, the alarm may be regarding a fault in a piece of equipment included in a building that has not generated an alarm before, or has not generated an alarm for a specified period of time (e.g., has not generated an alarm for one week or longer). In some examples, a condition of an alarm can include the alarm having recently occurred and is therefore being repeated. In some examples, a condition of an alarm can include the alarm expiring." See also col 4 ln 57 - col 5 ln 10. contextual teaches not generated for a specified period of time; contextual is a piece of equipment not generated an alarm before; activation signal issue state
Fuller then teaches identifying a set of alert disposition options for the first alert based on the alert type In col 4 ln 57 - col 5 ln 2: "Attributes of an alarm can include a priority of the alarm. In some examples, a priority of an alarm can include an urgent priority. For instance, the alarm may be regarding a fault in a piece of critical equipment included in a building, such as equipment included in a security system or HVAC system. In some examples, a priority of an alarm can include a high priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be important, but may not need to be addressed in an immediate fashion. In some examples, a priority of an alarm can include a low priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be ignored for a longer period of time than a fault corresponding to a high or urgent priority alarm."
Fuller then teaches presenting, through a user interface a list… of the plurality of alerts and alert disposition options in Fig 2 item 220.
Fuller then teaches recording the alert disposition option selected by a user for the first alert in col 4 ln 11-16: “Database 104 can include a number of suppressed alarms. For instance, database 104 can include a number of alarms suppressed by each of the users of the number of mobile devices 112-1, 112-2, 112-N, respectively. Each suppressed alarm can include instructions to suppress notifications of the alarm for a predetermined length of time.”
Fuller does not teach classifying an alert disposition option for the first alert using a classifier model, wherein the classifier model estimates a probability of use of the alert disposition option within the set of alert disposition options, the classifier model initially trained using historical alert data; determining an alert risk score for the first alert, wherein the alert risk score aggregates one or more risk model outputs, further wherein the one or more risk model outputs is based on an alert disposition option classification, a level of security interest of the alert disposition option, and a cost of loss of an asset monitored by the facility security system;
Beale teaches classifying an alert disposition option for the first alert using a classifier model, wherein the classifier model estimates a probability of use of the alert disposition option within the set of alert disposition options, the classifier model initially trained using historical alert data in col 5 ln 64 - col 6 ln 12: "In some embodiments, the learning module can identify a score to determine whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm. For example, the score can be indicative of a likelihood or a probability that the combination represents the false alarm or the valid alarm. In some embodiments, the score can be based on an amount by which the alarm signal and the additional information match the plurality of alarm signals from the historical time period and the plurality of additional information from the historical time period, and in some embodiments, the alarm signal and/or the additional information can be automatically or manually assigned different weights for such a matching comparison."
Beale then teaches determining an alert risk score for the first alert, wherein the alert risk score aggregates one or more risk model outputs, further wherein the one or more risk model outputs is based on an alert disposition option classification, of the alert disposition option, monitored by the facility security system in col 6 ln 27-39: " However, in some embodiments, the score can include a range of values with a calculated distribution (e.g. Gaussian) that indicates whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm. In such embodiments, the automated dispatcher module can include a cumulative distribution function that indicates when the automated dispatcher module should alert the user and/or the authorities, and in some embodiments, a sensitivity of the automated dispatcher module to the score can be automatically or manually adjusted based on the user preference data, such as days of the week or when the user is out of town.”
Beale then teaches a level of security interest in col 7 ln 16-20: “Additionally or alternatively, in some embodiments, a sensitivity of the central monitoring station to the score can be automatically or manually adjusted based on a price or level of service that the central monitoring station provides to the user.”
Beale then teaches prioritizing the first alert in col 6 ln 27-39: " However, in some embodiments, the score can include a range of values with a calculated distribution (e.g. Gaussian) that indicates whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm. In such embodiments, the automated dispatcher module can include a cumulative distribution function that indicates when the automated dispatcher module should alert the user and/or the authorities, and in some embodiments, a sensitivity of the automated dispatcher module to the score can be automatically or manually adjusted based on the user preference data, such as days of the week or when the user is out of town.
Beale then teaches and retraining the classifier model to estimate probability of use for the set of alert disposition options based on the alert disposition option selected by the user for the first alert in col 5 ln 11-19: "Furthermore, in some embodiments, the learning module can update the false alarm predicting model for increased accuracy at future times. For example, in some embodiments, the learning module can receive feedback signals indicating whether the combination of the alarm signal and the additional information represents the false alarm or the valid alarm and can use those feedback signals to update the false alarm predicting model for the increased accuracy at the future times."
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the alert risk score teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Fuller does not teach: automatically executing, for the first alert and based on the alert risk score of the first alert, a follow-up action with respect to a piece of building equipment associated with a sensor of the plurality of sensors corresponding to the first alert, the follow-up action comprising transmitting, via a communications interface of the facility security system, a control signal that causes the piece of building equipment to change an operational state
Trundle teaches automatically executing, for the first alert and based on the alert risk score of the first alert, a follow-up action with respect to a piece of building equipment associated with a sensor of the plurality of sensors corresponding to the first alert, the follow-up action comprising transmitting, via a communications interface of the facility security system, a control signal that causes the piece of building equipment to change an operational state in col 9 ln 26-51: "In the example shown in FIG. 1C, the control panel 20 detects a basement door opening event at 2:00 AM on a Monday based on receiving an indication from the basement door sensor 22. The control panel 20 provides an alert to the monitoring server 30 indicating the detected alarm event, and the monitoring server 30 proceeds to access data relevant to the detected alarm event. The monitoring server accesses data that includes data indicating (1) that there was no interior motion detected by the motion sensor 24, (2) that the alarm system monitoring the property 10 detected the phone line being cut prior to the basement door sensor 22 detecting the basement door being opened, (3) that historical sensor data collected from the alarm system indicates that the basement door is never opened around 2:00 AM, and (4) that there has been a recent spike in night time break-ins for the region of the property 10. The monitoring system 30 includes these factors in the alarm probability factors 40, where the monitoring system 30 then evaluates the factors to determine an alarm probability score estimating the likelihood that basement door opening at 2:00 AM on Monday should be treated as an emergency situation. Based on the evaluation, the monitoring server 30 determines an alarm probability score of 95% associated with the detected alarm event, and subsequently determines how to handle the detected alarm event based on the determined alarm probability score." See also col 11 ln 43-42: "The monitoring system control unit 110 communicates with the module 122 and the camera 130 to perform surveillance or monitoring. The module 122 is connected to one or more lighting systems and is configured to control operation of the one or more lighting systems. The module 122 may control the one or more lighting systems based on commands received from the monitoring system control unit 110. For instance, the module 122 may cause a lighting system to illuminate an area to provide a better image of the area when captured by a camera 130."
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify to modify the alert teachings of Fuller with the follow up action based on two alerts teaching of Trundle because Trundle teaches in col 1 ln 57-67: “False alarms are a significant issue for security systems. Security systems often detect events that reflect potential alarm conditions, but that are not in fact alarm situations. When a false alarm is detected by a security system and no one is available to confirm that the alarm is false, emergency services may be dispatched unnecessarily. Dispatching emergency services for false alarms consumes resources of emergency personnel and limits the ability of emergency personnel to respond to real alarm situations.” As Trundle’s teaching would save resources, one would be motivated to modify Fuller with Trundle so fewer resources would be consumed.
Fuller does not teach and a cost of loss of an asset monitored by the facility security system; prioritizing the first alert based on the alert risk score; presenting, through a user interface a prioritized list of the plurality of alerts, the prioritized list comprising the plurality of alerts, alert risk scores,
Pourmohammad teaches and a cost of loss of an asset monitored by the facility security system in par 0323: “Regardless of the imminent threat and its nature, the value of the asset is important in evaluating the risk to the owner. The asset value becomes more important when a company has multiple types of assets with different functionality and responsibilities. Some of them might be strategic and very valuable. But, others might be smaller and less valuable compared to the others. Asset assessment includes the asset cost estimation besides vulnerability assessment. The result of the asset value assessment is translated to a number between 1 to 10 in the risk model to represent the least to most valuable assets.”
Pourmohammad then teaches prioritizing the first alert based on the alert risk score in par 406: "Referring now to FIGS. 28-29, interfaces 2800 and 2900 are shown including a list of active threats for an asset listed along with the risk associated for each threat. The higher the risk score the more important that threat is."
Pourmohammad then teaches presenting, through a user interface a prioritized list of the plurality of alerts, the prioritized list comprising the plurality of alerts, alert risk scores, in par 406-409: "Referring now to FIGS. 28-29, interfaces 2800 and 2900 are shown including a list of active threats for an asset listed along with the risk associated for each threat. The higher the risk score the more important that threat is. In this regard, the monitoring client 128 can dynamically prioritize alarms, i.e., threats, based on a risk score associated with the asset affected by the threat. The monitoring client 128 can be configured to dynamically sort the threats of the list 2802 and 2902 so that the highest risk scores are shown on the top of the list, allowing a user to quickly identify what threats and/or assets are associated with the highest priority. As can be seen in interfaces 2800-2900, as new threats are reported, and risk scores change, threats can move up and down the list, as can be seen from list 2802 to 2902." and Figs 28-29.
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the cost of an asset in determining risk score teaching of Pourmohammad because Pourmohammad teaches in par 0211 that: “The analytics systems and methods as described herein can generate risk information for use in prioritization of alarms, presenting users with contextual threat and/or asset information, reducing the response time to threats by raising the situational awareness, and automating response actions.” As it would present useful information to users and reduce time to threats and automate response actions, one would be motivated to modify Fuller with Pourmohammad.
Per claim 12, Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claim 11, above. Fuller further teaches wherein the alert type is associated with an alert severity rating in col 4 ln 57 – col 5 ln 2: “Attributes of an alarm can include a priority of the alarm. In some examples, a priority of an alarm can include an urgent priority. For instance, the alarm may be regarding a fault in a piece of critical equipment included in a building, such as equipment included in a security system or HVAC system. In some examples, a priority of an alarm can include a high priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be important, but may not need to be addressed in an immediate fashion. In some examples, a priority of an alarm can include a low priority. For instance, the alarm may be regarding a fault in a piece of equipment that may be ignored for a longer period of time than a fault corresponding to a high or urgent priority alarm”
Per claim 13, Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claim 11, above. Fuller further teaches wherein the plurality of alerts further comprise a set of alert contextual data, the set of alert contextual data comprising a set of alert metadata, a set of threat data, and a set of facility data in col 5 ln 2-34: “Attributes of an alarm can include a source of the alarm. In some examples, the source of an alarm may include the source of the equipment with a fault. For instance, an alarm may be generated as a result of a fault in a piece of HVAC equipment, and the source of the fault may be a valve of a cooling coil of an air handling unit (AHU) is stuck open, a radiator valve is stuck shut, a hot water pump has stopped working, etc. In some examples, the source of an alarm may include the location of the equipment generating the fault. For instance, an alarm may be generated as a result of a fault in a piece of HVAC equipment, and the source of the fault may be the cooling coil of the AHU located on a third floor equipment room of a building, a radiator valve located in a first floor classroom of a building, or a hot water pump located in an underground boiler room of the building, etc.
Attributes of an alarm can include a category of the alarm. In some examples, the category of an alarm may include the category of the equipment with a fault. For instance, an alarm may be generated as a result of a fault in a piece of HVAC equipment, a fault in a security system, a fault in an electrical system, a fault in a plumbing system, etc. In some examples, the category of an alarm may include a Point Alarm (e.g., an alarm regarding a part of a piece of equipment), System Alarm (e.g., software or core issues such as an expired software license, etc.), and/or other categories of alarms.
Although attributes of the alarms are described as including a condition, priority, source, and category of the alarms, embodiments of the present disclosure are not so limited. For example, attributes of the alarms can include other alarm descriptors/characteristics.”
Fuller does not teach a set of environmental data .
Beale teaches a set of environmental data in col 3 ln 21-28: “In accordance with disclosed embodiments, the security system can protect a geographic area, and in some embodiments, the additional information can include weather data from a time associated with the alarm signal, movement data associated with the geographic area during the time associated with the alarm signal, a location of users of the security system during the time associated with the alarm signal, and/or incident reports relevant to the geographic area.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the set of environmental data teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Per claim 14, Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claim 11, above. Fuller further teaches wherein the set of alert disposition options for the alert type comprises a table of one or more actions that a user may select to dispose of the first alert, wherein the table is stored on one or more memory devices associated with the facility security system in col 6 ln 6-20: “In some examples, a user may have previously suppressed notifications of an alarm regarding a fault of a valve included in HVAC equipment; in response to computing device 106 determining that the attributes of another alarm regarding the fault of the valve received from building automation system 108 match those of the previously suppressed alarm regarding the fault of the valve, computing device 106 can refrain from transmitting a notification of the alarm regarding the fault of the valve to a mobile device among the number of mobile devices 112-1, 112-2, 112-N.” The user, selecting to suppressed, is a table of an action that a user may select as the action is in a database, which is a table.
Per claim 15, Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claim 14, above. Fuller further teaches wherein an option in the set of alert disposition options is assigned a code, wherein the code indicates a level of security significance in col 9 ln 1 – 16: “After a user has received the notification of alarm 216 and has considered actions to take regarding the alarm, the user can suppress further notifications for the alarm and/or notifications of other alarms with similar attributes to the alarm using suppress options 220. For example, a user can utilize suppress options 220 to suppress further notifications for the alarm and/or notifications of other alarms with similar attributes to the alarm for a predetermined length of time, as will be further described in connection with FIG. 3.
Once a user has elected to suppress notifications for the alarm and/or notifications of other alarms with similar attributes to the alarm using suppress options 220, a computing device (e.g., computing device 106, previously described in connection with FIG. 1) can receive the instructions to suppress notifications of the alarm and/or notifications of other alarms with similar attributes to the alarm.” The suppress options under a broadest reasonable interpretations teach a code because it can either suppress notification for the same alarm or for that alarm and similar alarms.
Per claim 19, Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claim 11, above. Fuller does not teach wherein the classifier model is trained to estimate the probability of use of an alert disposition option within the set of alert disposition options using a first data set
Beale teaches wherein the classifier model is trained to estimate the probability of use of an alert disposition option within the set of alert disposition options using a first data set in In col 6 ln 2-22: “For example, the score can be indicative of a likelihood or a probability that the combination represents the false alarm or the valid alarm. In some embodiments, the score can be based on an amount by which the alarm signal and the additional information match the plurality of alarm signals from the historical time period and the plurality of additional information from the historical time period, and in some embodiments, the alarm signal and/or the additional information can be automatically or manually assigned different weights for such a matching comparison. Furthermore, the learning module can transmit the score to the automated dispatcher module, for example, with the status signal. Then, the automated dispatcher module can compare the score to a threshold value to automatically determine whether to alert the user and/or the relevant authorities about the alarm signal. When such a comparison and/or the score indicates that the automated dispatcher module should alert the user and/or the relevant authorities, the automated dispatcher module can automatically alert the user and/or the relevant authorities about the alarm signal without human intervention.” See also col 5 ln 20-29: “In some embodiments, any of the feedback signals described herein can include user input explicitly identifying the alarm signal or the plurality of alarm signals from the historical time period as the valid alarm or the false alarm. Additionally or alternatively, in some embodiments, any of the feedback signals described herein can include information related to actions executed in response to the alarm signal or the plurality of alarm signals from the historical time period that are indicative of the valid alarm or the false alarm.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the alert risk score teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Per claim 20, Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claim 11, above. Fuller does not teach wherein the classifier model is retrained to estimate the probability of use of an alert disposition option within the set of alert disposition options using a second data set, the second data set comprising alert disposition codes applied by a user to alerts and alert contextual data .
Beale teaches wherein the classifier model is retrained to estimate the probability of use of an alert disposition option within the set of alert disposition options using a second data set, the second data set comprising alert disposition codes applied by a user to alerts and alert contextual data in In col 6 ln 2-22: “For example, the score can be indicative of a likelihood or a probability that the combination represents the false alarm or the valid alarm. In some embodiments, the score can be based on an amount by which the alarm signal and the additional information match the plurality of alarm signals from the historical time period and the plurality of additional information from the historical time period, and in some embodiments, the alarm signal and/or the additional information can be automatically or manually assigned different weights for such a matching comparison. Furthermore, the learning module can transmit the score to the automated dispatcher module, for example, with the status signal. Then, the automated dispatcher module can compare the score to a threshold value to automatically determine whether to alert the user and/or the relevant authorities about the alarm signal. When such a comparison and/or the score indicates that the automated dispatcher module should alert the user and/or the relevant authorities, the automated dispatcher module can automatically alert the user and/or the relevant authorities about the alarm signal without human intervention.” See also col 5 ln 20-29: “In some embodiments, any of the feedback signals described herein can include user input explicitly identifying the alarm signal or the plurality of alarm signals from the historical time period as the valid alarm or the false alarm. Additionally or alternatively, in some embodiments, any of the feedback signals described herein can include information related to actions executed in response to the alarm signal or the plurality of alarm signals from the historical time period that are indicative of the valid alarm or the false alarm.” Under a broadest reasonable interpretation this teaches “a second data set” as the second data set is defined by alert disposition codes applied by a user to alerts and alert contextual data. The user is applying valid or false and those teach under a broadest reasonable interpretation alert disposition codes. Information related to actions in response to the alarm signal teaches contextual data.
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the alert risk score teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Claim(s) 7 and16-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Fuller et al., US Pat. No. 9754478 B1 (“Fuller”) in view of Beale et al., US Pat No 10762773 B1 ("Beale"), further in view of Trundle US Pat No 9013294 B1 ("Trundle"), further in view of Pourmohammad et al., US PGPUB 20190138512 A1 (“Pourmohammad”), further in view of Schuster et al., US PGPUB 20190123931 A1 (“Schuster”).
Per claims 7 and 16, which are similar in scope, Fuller, Beale, Trundle, and Pourmohammad teach the limitations of claims 6 and 11, above. Fuller does not teach wherein the disposition probability is estimated by a machine learning model comprising one or more of a Bayesian network, a neural network, a state vector machine, a decision tree, a hidden Markov model, or a probabilistic relational model.
Schuster teaches building management system with vibration dataset. See abstract.
Schuster teaches wherein the disposition probability is estimated by a machine learning model comprising one or more of a Bayesian network, a neural network, a state vector machine, a decision tree, a hidden Markov model, or a probabilistic relational model in par 0102 and generating condition scores is taught in par 0103 which teaches disposition probability under a broadest reasonable interpretation.
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the machine learning model teaching of Schuster because Schuster teaches in par 034 that: “This approach to vibration data analytics provides several benefits and advantages associated with the BMS. For example, previous approaches have required manual initiation of the analytics process. In addition, these previous approaches have not used large amounts of historical data to fine-tune advanced building equipment models used to assess vibration data. Instead, they simply use rules-based approaches in order to detect problems. The automated analytics functionality described in the present disclosure drives increased insight, improved performance, longer lifetime, and efficient maintenance associated with building assets to name just a few examples.” As this would teach more effective maintenance of a building system one would be motivated to combine Fuller with Schuster.
Per claim 17, Fuller, Beale, Trundle, Pourmohammad, and Schuster teach the limitations of claim 16, above. Fuller does not teach wherein the alert risk scores are determined by a dynamic prioritization model based on inputs from or more of a contextual machine learning model, a database of historical alert data, alert contextual data, alert disposition data, a database of assets and asset costs, and a threat data service
Beale teaches wherein the alert risk scores are determined by a dynamic prioritization model based on inputs from or more of a contextual machine learning model, a database of historical alert data, alert contextual data, alert disposition data, a database of assets and asset costs, and a threat data service in col 3 ln 50-60: “In some embodiments, the learning module can build the false alarm predicting model by parsing historical data from a historical time period. For example, in some embodiments, the learning module can parse a plurality of alarm signals from the historical time period, a plurality of additional information from the historical time period, feedback signals indicative of a plurality of false alarms from the historical time period, and feedback signals indicative of a plurality of valid alarms from the historical time period to build the false alarm predicting model.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the database of historical alert teachings of Beale because Beale teaches in col 1 that: “Known security systems utilize a cloud server to process alarm signals and distribute the alarm signals to a central monitoring station for review and transmission of alert signals to users and/or relevant authorities when needed. However, known security systems often produce a high number of false alarms that consume bandwidth when transmitted and must be screened by live technicians at the central monitoring station, thereby greatly increasing costs associated with operating the central monitoring station.” As Beale’s teachings would reduce these costs one would be motivated to modify Fuller with Beale to make a less costly system.
Per claim 18 , Fuller, Beale, Trundle, Pourmohammad, and Schuster teach the limitations of claim 17, above. Fuller does not teach wherein the alert contextual data comprise internal contextual data and outputs of one or more machine learning models, the one or more machine learning models further comprising a spatial model, an occupancy model, a door classification model, and a sensor state model.
Pourmohammad teaches wherein the alert contextual data comprise internal contextual data and outputs of one or more machine learning models, the one or more machine learning models further comprising a spatial model, an occupancy model, a door classification model, and a sensor state model in par 0249: “The result of the enrichment by the asset information enricher 302 is the enriched threat 308. The enriched threat 308 can include an indication of a threat, an indication of an asset affected by the threat, and contextual information of the asset and/or threat. The RAP 120 includes risk engine 310 and risk score enricher 312. Risk engine 310 can be configured to generate a risk score (or scores) for the enriched threat 308. Risk engine 310 can be configured to generate a dynamic risk score for the enriched threat 308. The risk score enricher 312 can cause the dynamic risk can be included in the enriched threat 316 generated based on the enriched threat 308.” See also par 0317: “The risk engine 310 of the RAP 120 can be configured to generate risk scores for the threats via a model. The model used by the risk engine 310 can be based on Expected Utility Theory and formulated as an extended version of a Threat, Vulnerability and Cost (TVC) model. The risk engine 310 can be configured to determine the risk scores on a per asset basis. The threats can all be decoupled per asset in the processing pipeline as well as the calculation of the risk. For example, if a protest or weather condition is created alerts towards multiple buildings, separate alerts per building will be generated based on the geo-fences of the building and the detected alert.”
It would have been obvious to one ordinarily skilled in the art before the effective filing date of the claimed invention to modify the alert teachings of Fuller with the cost of an asset in determining risk score teaching of Pourmohammad because Pourmohammad teaches in par 0211 that: “The analytics systems and methods as described herein can generate risk information for use in prioritization of alarms, presenting users with contextual threat and/or asset information, reducing the response time to threats by raising the situational awareness, and automating response actions.” As it would present useful information to users and reduce time to threats and automate response actions, one would be motivated to modify Fuller with Pourmohammad.
Therefore, claims 1-20 are rejected under 35 USC 103.
Remarks:
Applicant argues:
Applicant's specification describes the associations between the equipment and the sensors, as well as several examples of building equipment distributed across a variety of building systems. For example, regarding the sensors, the Specification describes "[s]ensors 332 can include any type of sensor associated with a space or equipment within a building, and in particular can include sensors for use in building security" and "[f]or example, sensors 332 can include smoke detectors, temperature and humidity sensors, door position sensors, window position sensors, occupancy detectors, etc." at paragraph [0059].
Regarding building equipment, the Specification further explains "[r]emote systems and devices 334 can include any system or device that is not directly included within system 300, but that can be interfaced with system 300 to provide and/or receive data" and "[f]or example, remote systems and devices 334 can include a BMS for the building, an external or remote security system, an access control system, a surveillance system (e.g., including cameras and sensors), individual building equipment (e.g., controllers, fire safety devices, lighting components, etc.), or any other type of system or device" at paragraph [0059].
Claim 1 describes executing an automated follow-up action associated with a piece of
building equipment associated with a sensor. As described above with reference to paragraph [0059] of the Specification, such sensors may include various physical sensors to detect physical conditions within a building environment, and the building equipment may relate to physical building equipment. As such, Applicant submits this feature cannot be reasonably classified as part of the alleged mental process because the human mind cannot mentally execute such automated follow-up actions with respect to a piece of building equipment to cause the building equipment to change an operational state. Rather, this step of claim 1 is believed to qualify as an additional element beyond the alleged mental process.
Examiner responds:
The identified mental process is rejected because, as identified and similar to previous iterations of the claims, it is directed to a series of decision making steps or rules. The argument that these are “to detect physical conditions within a building environment” where the building equipment may refer to physical building equipment is a field of use statement, in other words, that this takes places within an environment is stating that the field of use is building. Further, that the building equipment may refer to physical building equipment includes equipment that is not physical building equipment. As the broadest reasonable interpretation is followed, the scope of building equipment includes any device, which could include a generic computer.
Applicant argues:
Moreover, the Federal Circuit in Enfish, LLC v. Microsoft Corp., 822 F.3d 1327 (Fed. Cir. 2016), held that claims directed to improvements in computer technology are not abstract. Here, claim 1 recites a specific technical improvement: a classifier model trained on historical alert disposition data that dynamically calculates disposition risk scores and alert risk scores to enable automated equipment control. These features improve how building security systems process and respond to alerts. This is not an abstract idea capable of being performed in the human mind, but rather a concrete technical implementation requiring specific computing resources and physical building infrastructure.
Examiner responds:
A “a concrete technical implementation requiring specific computing resources and physical building infrastructure” is not a computer improvement, but rather a series of steps that require computers and physical building resources. As was previously explained, the only improvement was non-technical in that a person ordinarily skilled in the art would not recognize this as an improvement in building security systems as none of the limitations technically limit security systems. This is a system of receiving, analyzing, and displaying the results of analysis that uses a security system, and one ordinarily skilled in the art would recognize that. No limitation the technical aspects of how a security system works; rather, the system describes how a system would be used. This is similar to applying a common business method to a computer, see MPEP 2106.05(f)(2), Alice, Versata, Gottschalk v Benson. Further, the limitations of the abstract idea are exchanging and analyzing information, which is patent ineligible subject matter, defined by cases such as Alice et seq. and Parker v Flook (alarm systems, specifically). See also, Electric Power Group. All of these cases are cited in the MPEP. Therefore this is unpersuasive. The only other system described in the spec is in fact a group of people who are performing these acts and therefore there is no technical improvement but rather using (applied use of) technology for what the people were tasked with doing. See specification par 004: “Since there may be many alerts and threat events, not only does the security platform require a large amount of resources, a high number of security operators and/or analysts may be required to review and/or monitor the various different alerts and threat events to assess risks posed to assets. Additionally, many alerts may be presented to security operators on an event by event basis without information to place the alert in situational context or to prioritize one alert over others.”
Applicant argues: To the extent that claim 1 is considered to recite any alleged abstract idea, Applicant submits that the additional elements described above integrate the alleged abstract idea into a
practical application for facilitating building security systems. The Federal Circuit has consistently held that claims which improve technological processes are not abstract. For example, in DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245 (Fed. Cir. 2014), the court found claims patent-eligible where they addressed a particular technological problem unique to the technological environment. Similarly, claim 1 addresses a technological problem specific to automated building security: efficiently prioritizing and responding to security alerts in real-time while controlling physical building equipment. The claimed solution-using a trained classifier model to estimate disposition probabilities, calculate risk scores, and automatically execute control signals to change equipment operational states-is rooted in the technology of building security automation and could not be performed outside of a computerized building security environment.
Examiner responds:
DDR is not persuasive as DDR described a technological solution to a problem necessarily rooted in computing. Here, and as well explained and illustrated in the previous rejection, Applicant’s alarm information limitations are in the non-specific area of dispatching, similar to 911 calls. While 911 calls use technology it is understood they have existed long before computers. Therefore the problems here are not necessarily rooted in computing.
Applicant argues:
Importantly, as the Federal Circuit recognized in McRO, Inc. v. Bandai Namco Games America Inc., 837 F.3d 1299 (Fed. Cir. 2016), claims that recite specific rules governing how data is processed to achieve an improved technological result are patent-eligible. Here, claim 1 recites specific rules for processing alert data: (1) estimating probabilities of use for disposition options using a trained classifier model; (2) calculating disposition risk scores using those probabilities; (3) calculating alert risk scores based on combinations of disposition risk scores; and (4) automatically executing control signals based on those risk scores. This ordered combination of specific processing rules transforms raw sensor alerts into prioritized, actionable security responses and provides a concrete technical improvement. Accordingly, Applicant submits that any alleged abstract idea recited in claim 1 is integrated into the practical application of securing a building and automatically performing or initiating follow-up actions with respect to operating states of building equipment based on identified security risks.
Examiner argues:
McRO is not persuasive as the evidence on record for McRO showed that the animation steps were those that formerly only people could do and this was replaced by rules for the animation system. There is no similar evidence here. Applicant’s ordered combination of specific processing rules is agreed with Examiner, this is the abstract idea. Transforming raw sensor alerts is unclear as the word raw is not in the claims, and unsure what the specific data processing transformation steps are in a technical sense. There is no specific data processing step only that data is processed in any and all ways by the generic elements recited by Applicant therefore this is unpersuasive.
Applicant argues:
Applicant emphasizes that the features of "automatically execute, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of building equipment of the building associated with a sensor of the plurality of sensors corresponding to the alert, the follow-up action comprising transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state" in claim 1 involves transmitting a control signal to change an operational state of building equipment. Additionally, the follow-up action is performed for an alert that is also presented based on its alert risk score, for example, due to the alert being a high-priority or high-risk alert, and the follow-up action is performed to address the alert. Accordingly, Applicant submits features including performing a follow-up-12-
action with respect to the building equipment in claim 1 cannot reasonably be characterized as generally applying the alleged judicial exception, but rather "appl[ies], rel[ies] on, or use[s] the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that the claim is more than a drafting effort designed to monopolize the judicial exception" (MPEP § 2106.04(d)) and therefore integrates the alleged judicial exception into a practical application.
Examiner responds:
This is not persuasive as the interpretation above shows that this would include merely having a computer do an information step and even if this were limited to what could be understood as a building component, like a security system (kept broad in the specification as a “system” therefore applicable to any and all systems and components of systems that could be a security system) would be more apply it limitations as an action is performed – a change in operational state, which is broad and encompasses any interaction with the system, therefore the element which has the operational state changed, broad in that it is operating in its ordinary capacity, has something happen due to the abstract idea and this is properly interpreted as apply it. MPEP 2106.05(f)(2). In sum, emphasis noted, change in operational state is apply it as nothing shows it is anything but something operating in its ordinary capacity and further this action is due to the steps of the mental process being applied to the additional element (building equipment, operational state).
Applicant argues:
Applicant also submits that the claimed invention recited in claim 1 provides "[a]n improvement in the functioning of a computer, or an improvement to other technology or technical field." (MPEP, 2106.04(d)(I)). In particular, the method of claim 1 reflects an improvement in the technical field of building security systems by automatically executing follow-up actions by causing building equipment to change an operational state, that can be performed according to determined alert risk scores related to alerts. For example, the building security system can use its limited resources (e.g., physical, computational, etc.) to adjust operational states for building equipment that are contributing to the most high-risk alerts. The Federal Circuit in Finjan, Inc. v. Blue Coat Systems, Inc., 879 F.3d 1299 (Fed. Cir. 2018), found claims patent-eligible where they provided a new kind of file that enables a computer security system to do things it could not do before. Analogously, the claimed invention provides a new approach to alert prioritization (i.e., using disposition code classifiers trained on historical data and modified TVC calculations) that enables building security systems to perform intelligent, automated responses that were not previously possible. The classifier model's ability to learn from historical disposition patterns and dynamically adjust risk scoring represents a technological improvement over conventional alert systems that rely on static rules or manual operator intervention.
With regards to the technical improvement, the MPEP states "The improvement can be provided by one or more additional elements...In addition, the improvement can be provided by the additional element(s) in combination with the recited judicial exception...Thus, it is important for examiners to analyze the claim as a whole when determining whether the claim provides an improvement to the functioning of computers or an improvement to other technology-13-
or technical field." (MPEP 2106.05(a), emphasis added). Applicant respectfully submits that the improvement in claim 1 is provided by the additional element of "automatically execute, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of building equipment of the building associated with a sensor of the plurality of sensors corresponding to the alert, the follow-up action comprising transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state" in combination with the previous claim elements when claim 1 is considered as a whole
Additionally, Applicant respectfully submits that the claimed classifier model trained on historical alert disposition data provides a concrete technical improvement that cannot be dismissed as mere data gathering or conventional machine learning. The classifier model operates in a specific technical manner: it is trained based on "selected alert disposition options for a plurality of historical alerts related to the building," then uses that training to "estimate probabilities of use for the set of alert disposition options." These probability estimates are then used to calculate "alert disposition risk scores" and "alert risk scores" through a multi-step algorithmic process. This is not generic machine learning applied to any data, but rather is a specialized classifier architecture designed for the specific technical problem of security alert prioritization in building environments. The Federal Circuit in Amdocs (Israel) Ltd. v. Openet Telecom, Inc., 841 F.3d 1288 (Fed. Cir. 2016), found claims patent-eligible where they solved a technological problem through a particular arrangement of technical components. Here, the specific arrangement of sensor inputs, disposition classifiers, risk score calculations, and automated equipment control provides an unconventional technological solution to the technical challenge of intelligent building security automation.
Examiner responds:
The arguments are carefully reviewed but a technical improvement was not recited in the claims. As stated in MPEP 2106.05(a),
“If it is asserted that the invention improves upon conventional functioning of a computer, or upon conventional technology or technological processes, a technical explanation as to how to implement the invention should be present in the specification. That is, the disclosure must provide sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement. The specification need not explicitly set forth the improvement, but it must describe the invention such that the improvement would be apparent to one of ordinary skill in the art.”
To clarify, as the claims are little changed and the rejection maintained under 101, no technical improvement was or is claimed. Applicant’s only specification argument that there is some kind of technical or patent eligible improvement is that improvements are for “decision making of security center operators.” These are people, not technical elements. So, the improvement is for how people work together. This is not supported by the MPEP quotes, as Applicant’s argument is that people are working better together and “resources” (ie, human labor or monies spent on human labor) are saved. Though this argument has been carefully considered, Applicant’s application to help people to work more efficiently together is not patent eligible subject matter; is not a technical improvement in terms of patent law; the rejection is proper; and the arguments made are not persuasive. Changing an operational state and taking information from sensors is not a technical improvement but at best applied use of tech to the abstract idea as one could view information from sensors and note the sensor information and changing an operational state could be displaying information on a screen as a device is included in what the operational state could be and the change is not further specified in the claims.
Amdocs argument is unpersuasive as the technical elements alone, in combination, and considering the claims as a whole, are applied to the abstract idea. The classifier elements are unchanged from the previous action and therefore the arguments are also unchanged. The classifier model is not claimed in a “specific technical manner” but rather is only claimed in terms of its desired functional result, see MPEP 2106.05(f)(1). There is no specialized architecture described for the classifier, this is at best an attorney interpretation and has no evidence. Where is the specific architecture that would distinguish this classifier from another classifier? Is it in the specification? If it isn’t then this classifier teaches any and all classifiers that perform the desired functional results, and therefore it is not specific but generic.
Applicant argues:
Applicant respectfully submits that claim 1 is also eligible under Step 2B as reciting
significantly more than an abstract idea. Regarding Step 2B, the MPEP states that, "Examiners should answer this question by first identifying whether there are any additional elements (features/limitations/steps) recited in the claim beyond the judicial exception(s), and then evaluating those additional elements individually and in combination to determine whether they contribute an inventive concept (i.e., amount to significantly more than the judicial exception(s))." (MPEP 2106.5(II), emphasis added).
Applicant submits that claim 1 recites a "non-conventional and non-generic arrangement of components" (MPEP, 2106.05) which amounts to significantly more than the alleged abstract idea. Even if some elements of claim 1 could be interpreted as abstract/mental steps as suggested in the Office Action, Applicant submits the feature of "automatically execute, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of building equipment of the building associated with a sensor of the plurality of sensors corresponding to the alert, the follow-up action comprising transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state" in claim 1 includes additional elements which amount to significantly more than the alleged abstract idea when considered in combination with the other features of claim 1.
As the Federal Circuit recognized in BASCOM Global Internet Services, Inc. v. AT&T Mobility LLC, 827 F.3d 1341 (Fed. Cir. 2016), an inventive concept can be found in the ordered combination of claim elements, even if those elements are individually conventional. Here, the ordered combination of: (1) receiving alerts from a plurality of physical sensors; (2) using a classifier model trained on historical disposition data to estimate probabilities; (3) calculating disposition risk scores and alert risk scores through specific algorithmic processes; and (4) automatically transmitting control signals to change equipment operational states based on those scores, represents a non-conventional arrangement that provides significantly more than merely implementing an abstract idea on generic computer hardware.
Many of the factors discussed above with respect to Step 2A Prong Two are relevant to Step 2B and such arguments are reiterated here with respect to Step 2B. In addition, Applicant submits that the additional element of "automatically execute, for an alert of the two or more of the plurality of alerts and based on an alert risk score of the alert, a follow-up action with respect to a piece of building equipment of the building associated with a sensor of the plurality of sensors corresponding to the alert, the follow-up action comprising transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state" in claim 1 amounts to significantly more than the alleged abstract idea because it contributes to the inventive concept when considered in combination with the preceding steps of claim 1. For example, previous features of claim 1 recite an improved use for a building security system that prioritizes alerts based on determined alert risk scores, and transmits control signals that causes a piece of building equipment to change an operational state. The features of performing any of the specified follow-up actions in claim 1 include "transmitting, via a communications interface of the building security system, a control signal that causes the piece of building equipment to change an operational state" that is "based on an alert risk score of the alert" and include particular actions performed based on how the alert risk score is calculated. As a whole, claim 1 determines an appropriate action to perform and then executes the appropriate action with respect to building equipment based on its determined alert risk score as claimed. Significantly, the claims do not preempt all methods of security alert prioritization or building automation. Rather, they claim a specific technical approach using disposition code classifiers, probability estimation, and multi-factor risk scoring to drive automated equipment control, leaving open numerous alternative approaches for achieving similar goals.
Applicant submits that these features amount to significantly more than an abstract idea and cannot reasonably be considered "generic and well understood" as asserted in the Office Action. The specific combination of disposition code classification, probability-based risk scoring, and automated equipment control responsive to those scores is not routine or conventional in the building security field. Indeed, the Specification describes how conventional-16-
systems suffered from deficiencies that the claimed invention addresses, namely the inability to dynamically prioritize alerts based on learned disposition patterns and to automatically execute appropriate follow-up actions without human intervention. See, e.g., Specification at [0003]- [0004], [0031], [0033], [0036], [0049], and [0064].
The Federal Circuit has cautioned that "[w]hether something is well-understood, routine, and conventional to a skilled artisan at the time of the patent is a factual determination." Berkheimer v. HP Inc., 881 F.3d 1360, 1369 (Fed. Cir. 2018). Here, the Office Action has not established with evidentiary support that the specific ordered combination of elements in claim 1 was well-understood, routine, or conventional at the time of filing. To the contrary, the lack of prior art rejections in the Office Action suggests that the Office considers the claimed invention to be not well-understood, routine, or conventional at the time of filing. Accordingly, Applicant submits that claim 1, when considered as a whole, amounts to significantly more than any alleged judicial exception.
Applicant submits that the use of the "classifier model" or "machine learning model" recited in the pending claims amounts to more than mere data gathering or conventional computer implementation. The claims do not recite generic machine learning. Rather, they recite a specific classifier model trained on a specific type of data (historical alert disposition options) for a specific purpose (estimating probability of use for alert disposition options). The probability estimates are then used in a specific multi-step process to calculate risk scores that drive automated physical actions (i.e., transmitting control signals to change building equipment operational states). This is not an abstract use of machine learning, but a concrete technical application that produces physical results in the real world. As the Federal Circuit recognized in Visual Memory LLC v. NVIDIA Corp., 867 F.3d 1253 (Fed. Cir. 2017), claims that recite a specific structure for achieving an improvement in computer functionality are not abstract, even if they involve data processing.
Examiner responds:
These arguments are considered but because the rejection does not find whether or not there is well-understood, routine, and conventional activity, but rather that the additional elements alone, in combination, and considering the claims as a whole, are apply it limitations, the arguments are not related to the rejection. In sum, as Applicant is unable to provide persuasive support for finding the generic, broad scope additional elements are not apply it, Applicant has not overcome the step 2B rejections. In the same section as was cited by applicant, MPEP 2106.05(II), the instructions are to carry over considerations from Prong 2 if from MPEP 2106.05(f): “• Carry over their identification of the additional element(s) in the claim from Step 2A Prong Two;
• Carry over their conclusions from Step 2A Prong Two on the considerations discussed in MPEP §§ 2106.05(a) - (c), (e) (f) and (h)”
As this was done, any argument that would be persuasive would have to show how the elements were not apply it, and this was already answered above. Therefore, the rejection under step 2B is maintained.
Examiner in furthering compact prosecution reviewed the specification again and did not find suggestions to overcome the rejection.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RICHARD W. CRANDALL whose telephone number is (313)446-6562. The examiner can normally be reached M - F, 8:00 AM - 5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Anita Coupe can be reached at (571) 270-3614. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/RICHARD W. CRANDALL/Primary Examiner, Art Unit 3619
Prosecution Insights