DETAILED ACTION
This action is responsive to the application filed on 03/16/2023. Claims 1-11 and 13-20 are pending and have been examined. This action is Final.
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 .
Priority
Applicant’s claim for the benefit of a prior-filed application under 35 U.S.C. 119(e) or under 35 U.S.C.
120, 121, 365(c), or 386(c) is acknowledged.
Response to Arguments
Argument 1: The applicant argues on pages 1-2 that the amended claims are patent eligible because they allegedly recite a specific technical improvement to a computer-implemented machine-learning process, rather than merely applying an abstract idea on generic computer components. The applicant points to the amended claim structure involving a plurality of machine learning models, prediction results, semantic information, correlations, threshold evaluation, selection of actionable outputs, and execution of a subset of executable actions before the event occurs. The applicant asserts that this claimed process improves how machine-learning prediction systems process and translate complex multi-parameter prediction outputs into domain-specific actionable results. The applicant relies on the specification’s examples, including preparing users for urgent or unmet needs before natural events and predicting changes in webpage visits or search volume so advertising strategies can be adjusted. The applicant analogizes the claims to Enfish and cites Ex-parte Desjardins to argue that the claims improve machine-learning processing itself, rather than merely applying an abstract idea downstream.
Response to Argument 1: The applicant's arguments regarding subject matter eligibility have been considered but are not persuasive. The amended claims do not recite a specific improvement to computer functionality or to the operation or training of a machine learning model itself. Rather, the claims use generic computer components, a plurality of machine learning models, prediction results, semantic information, correlations, thresholds, selection of actionable outputs, and execution of actions as tools to collect prediction information, evaluate and correlate that information with contextual information, select outputs, and execute actions based on the selected outputs. The claims do not recite a particular machine learning architecture, training technique, loss function, parameter update, data structure, or other technical mechanism that improves computer operation or improves machine learning processing itself. Instead, any alleged improvement is to the result of the abstract decision making process, namely deciding what actionable output should be generated, selected, or executed before a future real world event occurs. Accordingly, the rejection remains because the claims do not improve the computer or machine learning model itself, but merely apply computer and machine learning tools to generate and execute action based decisions using predicted event information and contextual information.
Argument 2: The applicant argues on pages 3-5 that the amended independent claims overcome the prior-art rejections because Mukherjee does not disclose or suggest the newly added requirement of obtaining prediction results from a plurality of machine learning models, where each machine learning model is associated with a respective event type and configured to generate a likelihood of that respective event type occurring in a geospatial region. The applicant specifically addresses the Office’s reliance on Mukherjee paragraphs 14 and 26, arguing that those portions only disclose a weather intelligence system using an AI model to predict whether a geographic region will experience an unusual weather condition, such as being hotter or colder than regular weather conditions. According to the applicant, this is not the same as multiple event-type-specific machine learning models. The applicant further argues that Appel, Li, and Spiegel do not cure this deficiency, and that independent claims 14 and 20, plus the dependent claims, are allowable for the same reasons.
Response to Argument 2: The applicant's arguments regarding the prior art have been considered but are not persuasive because the updated rejection is not based on Mukherjee alone for the newly added plurality of machine learning models limitation. Mukherjee is relied upon for the weather intelligence method that retrieves weather forecast data for geographic regions, determines geographic regions predicted to experience a weather anomaly or unusual weather condition, correlates those predicted geographic region outputs with business rule, user profile, consumer process, and offer information, and programmatically enables triggers to transmit service related offers to user devices. Watt is relied upon for training a plurality of different machine learning models for environmental and weather related prediction in geospatial and spatiotemporal contexts. Appel is relied upon for threshold based selection or generation of mitigation recommendations when a future event related score satisfies a threshold. High is relied upon for computer executed processing of upcoming events, matching detected events to categories, retrieving corresponding database information, and executing computer directed actions before the future event occurs. Further, with respect to claims 2 and 15, Li is additionally relied upon for the claimed probability or confidence based prediction information, as set forth in the claim mappings. Thus, even if Mukherjee alone does not disclose each machine learning model of a plurality of machine learning models being associated with a respective event type and configured to generate a likelihood of that respective event type occurring in a geospatial region, the combination of Mukherjee, Watt, Appel, High, and Li teaches or renders obvious the amended claim structure when the references are considered as a whole. Therefore, the amendments do not overcome the updated prior art rejection of independent claims 1, 14, and 20, or dependent claims 2 and 15, and the remaining dependent claims remain unpatentable for the reasons set forth in the claim mappings.
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 1-11 and 13-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding claim 1,Step 1: This claim is directed to a computer-implemented method, which is a process, one of the four statutory categories. Therefore, claim 1 satisfies Step 1.
Step 2A Prong 1:(a) “A computer-implemented method for utilizing predictions of future real-world events to make actionable decisions that prepare users for the real-world events, comprising: correlating... one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region; selecting... a subset of the one or more actionable outputs whose correlations satisfy a threshold value;” -- The limitation is directed to a method for utilizing predictions for real-world events and making decisions for users by correlating prediction results to semantic information and selecting outputs based on whether correlations satisfy a threshold value. The limitation recites evaluation, observation, and judgment that can be practically performed in the human mind, with aid of pen and paper. The limitation also recites mathematical concepts, including correlations and threshold comparisons. Thus, the limitation is directed to an abstract idea, including mental processes and mathematical concepts.
Step 2A Prong 2 and Step 2B:(a) “obtaining, by a computer and from a plurality of machine learning models, a plurality of prediction results in association with an event, wherein each machine learning model of the plurality of machine learning models is associated with a respective event type and is configured to generate a likelihood of a respective event type occurring in a geospatial region; obtaining, by the computer and from a database, semantic information;” -- The limitation recites obtaining prediction results and obtaining semantic information using generic computer components, machine learning models, and a database. The limitation is directed to data gathering and receiving information for use in the abstract correlation and decision-making process. The claim does not recite a particular machine learning architecture, a particular training technique, a particular loss function, a particular parameter update, or any technical improvement to the machine learning models or computer functionality. Therefore, the limitation amounts to insignificant extra-solution activity and does not integrate the judicial exception into a practical application (see MPEP 2106.05(g)). Furthermore, under Step 2B, obtaining data using generic computer components, machine learning models, and databases is well-understood, routine, and conventional activity and does not provide significantly more than the judicial exception (see MPEP 2106.05(d)(II)).
(b) “generating, by the computer for a user device, one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information;” -- The limitation recites generating outputs based on the results of the abstract correlation and decision-making process. The claim does not recite a particular technical mechanism for generating the outputs, nor does the claim recite an improvement to computer functionality or another technology. Therefore, the limitation amounts to insignificant extra-solution activity and does not integrate the judicial exception into a practical application (see MPEP 2106.05(g)). Furthermore, under Step 2B, generating output data based on processed information is well-understood, routine, and conventional activity and does not provide significantly more than the judicial exception (see MPEP 2106.05(d)(II)).
(c) “executing, by the computer, the subset of executable actions associated with the respective event type prior to occurrence of the event occurring in the geospatial region…by a computer... from a plurality of machine learning models... by the computer... from a database... by the computer for a user device” -- The limitation recites executing unspecified executable actions associated with the event type before the event occurs, and merely applies the recited steps using generic computer components, machine learning models, a database, and a user device. The limitation merely applies the result of the abstract data analysis and decision-making process in a generic computer environment. Therefore, the limitation amounts to no more than mere instructions to apply the abstract idea using a computer and does not integrate the judicial exception into a practical application or provide significantly more than the judicial exception (see MPEP 2106.05(f)).
Thus, claim 1 is non-patent eligible. Claims 14 and 20 are analogous to claim 1 aside from claim type and minimal changes, and therefore the same rejection can be applied.
Regarding claim 2,
Step 1: This claim is directed to a computer-implemented method (a process), which is one of the four statutory categories. Therefore, claim 2 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 1, wherein each of the plurality of prediction results comprises one or more parameters of: (1) the event; (2) a time window that the event occurs; (3) a geospatial area that the event occurs; (4) an intensity of the event; and (5) a probability of occurrence of the event corresponding to one or more of (2), (3), and (4).” -- The limitation recites prediction results will further comprise a list of parameters recited in the claim. The claim does not amount to no more than mere further limiting to a field of use/environment, without any integration to a practical application, nor significantly more than the judicial exception (see MPEP 2106.05(h)).
Thus, claim 2 is non-patent eligible. Claim 15 is analogous to claim 2, aside from claim type, and thus the same rejection will apply.
Regarding claim 3,Step 1: This claim is directed to a computer-implemented method, which is a process, one of the four statutory categories. Therefore, claim 3 satisfies Step 1.
Step 2A Prong 1:(a) “The computer-implemented method of claim 1, wherein correlating, by the computer, the one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region is based on a predetermined mapping relationship between the plurality of prediction results and the semantic information.” -- The limitation is directed to correlating prediction results of a respective event type to semantic information for a geospatial region based on a predetermined mapping relationship between prediction results and semantic information. The limitation recites associating one set of information with another set of information based on a predetermined relationship, which can be performed in the human mind using evaluation, observation, and judgment, with aid of pen and paper. Thus, the limitation is directed to an abstract idea, including a mental process.
There are no elements to be evaluated under Step 2A Prong 2 and Step 2B.
Thus, claim 3 is non-patent eligible. Claim 16 is analogous to claim 3, aside from claim type, and thus the same rejection will apply.
Regarding claim 4,Step 1: This claim is directed to a computer-implemented method, which is a process, one of the four statutory categories. Therefore, claim 4 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 1, wherein generating, by the computer for the user device, one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information comprises: providing the one or more prediction results and the semantic information to a trained machine learning model; and receiving, from the trained machine learning model, an output including data that identifies the one or more actionable outputs.” -- The limitation recites generating actionable outputs by providing prediction results and semantic information to a trained machine learning model and receiving an output from the trained machine learning model. The claim does not recite a particular machine learning architecture, a particular training technique, a particular loss function, a particular parameter update, or any technical improvement to the trained machine learning model or computer functionality. The limitation amounts to insignificant extra-solution activity because it merely provides information to a generic trained machine learning model and receives output data from that model for use in the abstract decision-making process, and it cannot be integrated to a practical application (see MPEP 2106.05(g)). Furthermore, under Step 2B, providing data to a machine learning model and receiving output data from the machine learning model is well-understood, routine, and conventional activity (WURC) and does not provide significantly more than the judicial exception (see MPEP 2106.05(d)(II)).
Thus, claim 4 is non-patent eligible. Claim 17 is analogous to claim 4, aside from claim type, and thus the same rejection will apply.
Regarding claim 5,
Step 1: This claim is directed to a computer-implemented method (a process), which is one of the four statutory categories. Therefore, claim 5 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 1, wherein the one or more actionable outputs is automatically executed by the user device.” -- The limitation recites that the actionable outputs is automatically executed by the user device (computer), and does not integrate to a practical application and does not provide significantly more than the judicial exception (see MPEP 2106.05(f)).
Thus, claim 5 is non-patent eligible. Claim 18 is analogous to claim 5, aside from claim type, and thus the same rejection will apply.
Regarding claim 6,
Step 1: This claim is directed to a computer-implemented method (a process), which is one of the four statutory categories. Therefore, claim 6 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 1, wherein the one or more actionable outputs is displayed by the user device to a user using a graphic user interface (GUI).” -- The limitation recites that the output will be displaying by the user device to a user and also using a GUI. The limitation amounts to no more than mere instructions to apply onto a computer and it does not integrate to a practical application nor provides significantly more than the judicial exception (see MPEP 2106.05(f)).
Thus, claim 6 is non-patent eligible. Claim 19 is analogous to claim 6, aside from claim type, and thus the same rejection will apply.
Regarding claim 7,
Step 1: This claim is directed to a computer-implemented method (a process), which is one of the four statutory categories. Therefore, claim 7 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 1, wherein the event comprises a natural event or condition.” -- The limitation recites that the event will further comprise natural event or condition. The limitation amounts to no more than further limiting to a field of use/environment and it does not integrate to a practical application, nor provide significantly more than the judicial exception (see MPEP 2106.05(h)).
Thus, claim 7 is non-patent eligible.
Regarding claim 8,
Step 1: This claim is directed to a computer-implemented method (a process), which is one of the four statutory categories. Therefore, claim 8 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 7, wherein the natural event or condition comprises a natural hazardous event.” -- The limitation recites that the natural event/condition will further comprise natural hazardous event. The limitation amounts to no more than further limiting to a field of use/environment and it does not integrate to a practical application, nor provide significantly more than the judicial exception (see MPEP 2106.05(h)).
Thus, claim 8 is non-patent eligible.
Regarding claim 9,
Step 1: This claim is directed to a computer-implemented method (a process), which is one of the four statutory categories. Therefore, claim 9 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 1, wherein the semantic information comprises information corresponding to a product, a service, and/or a provider thereof” -- The limitation recites that the semantic information that comprises product, service, and a provider-corresponded information. The limitation amounts to no more than mere further limiting to a field of use/environment, and it does not integrate to a practical application, nor provide significantly more than the judicial exception (see MPEP 2106.05(h)).
Thus, claim 9 is non-patent eligible.
Regarding claim 10,
Step 1: This claim is directed to a computer-implemented method (a process), which is one of the four statutory categories. Therefore, claim 10 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
“The computer-implemented method of claim 1, wherein obtaining, by the computer and from the plurality of machine learning models, the plurality of prediction results in association with the event is based on the semantic information.” -- The limitation recites using the computer to obtain prediction results associated with an event based on semantic information. The limitation is directed to an insignificant, extra-solution activity that cannot be integrated to a practical application (see MPEP 2106.05(g)). Furthermore, under Step 2B, the act of obtaining data based on gathered information to be implemented onto the computer is a well-understood, routine, and conventional activity (WURC) that cannot provide significantly more than the judicial exception (see MPEP 2106.05(d)(II)).
Thus, claim 10 is non-patent eligible.
Regarding claim 11,Step 1: This claim is directed to a computer-implemented method, which is a process, one of the four statutory categories. Therefore, claim 11 satisfies Step 1.
Step 2A Prong 1:(a) “The computer-implemented method of claim 1, wherein generating the one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information is in response to determining that at least one of the one or more parameters of the event satisfies a predetermined threshold.” -- The limitation is directed to generating actionable outputs based on correlations of the prediction results and semantic information in response to determining whether an event parameter satisfies a predetermined threshold. The limitation recites evaluating information, comparing a parameter to a threshold, and generating outputs based on that determination. Such threshold comparison and evaluation can be performed in the human mind using observation and judgment, with aid of pen and paper, and also recites a mathematical concept in the form of a threshold determination. Thus, the limitation is directed to an abstract idea, including mental processes and mathematical concepts.
There are no additional elements to be evaluated under Step 2A Prong 2 and Step 2B.
Thus, claim 11 is non-patent eligible.
Regarding claim 13,
Step 1: This claim is directed to a computer-implemented method, which is a process, one of the four statutory categories. Therefore, claim 13 satisfies Step 1.
There are no elements to be evaluated under Step 2A Prong 1.
Step 2A Prong 2 and Step 2B:
(a) “The computer-implemented method of claim 1, wherein correlating, by the computer, the one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region is by using a machine learning model trained on historical events and historical data in correspondence to the semantic information.” -- The limitation recites performing the correlating step by using a machine learning model trained on historical events and historical data corresponding to semantic information. The claim does not recite a particular machine learning architecture, a particular training technique, a particular loss function, a particular parameter update, or any technical improvement to the machine learning model or computer functionality. Rather, the limitation merely applies the abstract correlation of prediction results and semantic information using a generic trained machine learning model. Therefore, the limitation amounts to no more than mere instructions to apply the abstract idea on a computer and does not integrate the judicial exception into a practical application or provide significantly more than the judicial exception (see MPEP 2106.05(f)).
Thus, claim 13 is non-patent eligible.
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.
Claims 1, 3-11, 13-14, 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over US20210215848A1, by Mukherjee et. al. (referred herein as Mukherjee) in view of US20210256378A1, by Watt et. al. (referred herein as Watt) in view of US20200134545A1, by Appel et. al. (referred herein as Appel) further in view US20170278053A1, by High et. al. (referred herein as High).
Regarding claim 1, Mukherjee teaches:
A computer-implemented method for utilizing predictions of future real-world events to make actionable decisions that prepare users for the real-world events, comprising: ([Mukherjee, page 1] “A weather intelligence system retrieves weather forecast data for a number of geographic regions. The weather intelligence system determines, using the weather forecast data for each of the geographic regions, a set of geographic regions predicted to experience a weather anomaly, or unusual weather condition, during a particular time interval…The weather intelligence system can then programmatically enable a trigger to transmit a service - related offer associated with the weather anomaly to user devices located within one of the geographic regions predicted to experience a weather anomaly, or unusual weather condition, during a particular time interval.”, wherein the examiner interprets a weather intelligence system that retrieves weather forecast data and determines geographic regions predicted to experience a weather anomaly and programmatically enables a trigger to transmit a service-related offer to user devices to be the same as a “computer-implemented method for utilizing predictions of future real-world events to make actionable decisions that prepare users for the real-world events” because both are directed to systems that use predictive data about future conditions to generate and deliver actionable information to users before those conditions occur)
obtaining, by a computer and from a machine learning model, a plurality of prediction results in association with an event; ([Mukherjee, [0026] “Using the weather forecast data, the weather intelligence system 120 runs an AI model to predict whether a geographic region will experience an unusual weather condition, such as being hotter or colder than its regular weather conditions. The AI model learns the short-term weather data patterns over the last few days at a given time and makes predictions based on the learning” AND [Mukherjee, [0014]] “The weather intelligence system determines , through a machine - learning process , whether the weather forecast data indicates conditions that people would generally consider to be unusually hot or cold. The weather intelligence system can then programmatically enable a trigger to transmit a service - related offer associated with the unusual weather condition to user devices located within one of the geographic regions where that weather condition is determined”, wherein the examiner interprets “the weather intelligence system 120 runs an AI model to predict whether a geographic region will experience an unusual weather condition” and “determines, through a machine-learning process, whether the weather forecast data indicates conditions” as well as “intelligence system determines , through a machine - learning process , whether the weather forecast data indicates conditions that people would generally consider to be unusually hot or cold” to be the same as “obtaining, by a computer and from a machine learning model, a plurality of prediction results in association with an event” because both are directed to a computer system using artificial intelligence or machine learning to generate multiple predictions about future conditions or events.)
obtaining, by the computer and from a database, semantic information; ([Mukherjee, [0033]] “The platform 10 can include various databases to store data used in the network computer system 100 and the weather intelligence system 120. These databases can include a configuration database 102 to store parameters, business rules, etc., an offer database 104 to store text, media files, and other information related to offers that can be sent to users, and a user database 106 to store profiles of users.” AND [Mukherjee, [0045]] “A rules engine 280 can then index user profiles from a database of user profiles 284 to determine which users registered with the environment are currently located within (or near) any of the geographic regions on the list. The rules engine 280 can also filter the user profiles by one or more programmed business rules 282, which enterprise customers can configure to target desired characteristics for offer recipients.”, wherein the examiner interprets parameters, business rules, offer information, and user profile data obtained from databases to be the same as “semantic information” because they are meaning-bearing contextual information used to determine what offer/output/action should be provided for a particular predicted condition and recipient context.)
correlating, by the computer, one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region; ([Mukherjee, 0026] “Using the weather forecast data, the weather intelligence system 120 runs an AI model to predict whether a geographic region will experience an unusual weather condition, such as being hotter or colder than its regular weather conditions.” AND [Mukherjee, [0035]] “In one aspect, a consumer process 250 registers to sign up for the weather intelligence service 200 in order to receive, on a regular schedule, a list of geographic regions experiencing a specific type of unusual weather condition” AND [Mukherjee, [0035]] “ In addition, this also helps the service provider to match the right set of zip codes to the right consumer process.” AND [Mukherjee, [0036]] “Prior to the handshake process, a registration scheduler 260 receives configuration data, including the type of weather trigger (e.g., unusually hot, unusually cold, or inclement) that consumer process 250 should be configured to listen to and sign up credentials for the trigger.” AND [Mukherjee, [0045]] “A consumer interface 270 receives the appropriate list of geographic regions from the service provider and creates corresponding weather-related triggers for each of the geographic regions on the list…A rules engine 280 can then index user profiles from a database of user profiles 284 to determine which users registered with the environment are currently located within (or near) any of the geographic regions on the list.”, wherein the examiner interprets matching weather trigger types/geographic regions/zip codes to consumer processes, user profiles, business rules, and offer information to be the same as “correlating, by the computer, one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region” because both are directed to associating predicted event information for a location or region with contextual database information to determine relevant downstream outputs.)
generating, by the computer for a user device, one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information; ([Mukherjee, [0045]] “For example, a retail location may configure an offer to send a push notification offering a discount on cold drinks to user devices of registered users within an unusually hot geographic region, but only for users who are not already regular customers of the business.” AND [Mukherjee, [0056]] “Once the process has filtered users that match the business rules, offers are sent to the appropriate users (460). In some implementations, offers can cause a user device to display the offer on a user interface of an application (e.g., through the user of a push notification) registered with the platform environment.”, wherein the examiner interprets sending offers or push notifications to user devices based on weather-trigger/geographic-region/user-profile/business-rule matching to be the same as “generating, by the computer for a user device, one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information” because both are directed to producing outputs for user devices based on correlations between predicted event conditions and contextual information.)
Mukherjee does not teach wherein each machine learning model of the plurality of machine learning models is associated with a respective event type and is configured to generate a likelihood of a respective event type occurring in a geospatial region…selecting, by the computer, a subset of the one or more actionable outputs whose correlations satisfy a threshold value; and executing, by the computer, he subset of executable actions associated with the respective event type prior to occurrence of the event occurring in the geospatial region.
Watt teaches wherein each machine learning model of the plurality of machine learning models is associated with a respective event type and is configured to generate a likelihood of a respective event type occurring in a geospatial region; ([Watt, Abstract] “A machine learning architecture is proposed that is directed to receive different time-series data sets relating to environmental conditions as well as a target variable for prediction and to transform the time-series data sets for training a plurality of different machine learning models.” AND [Watt, [0005]] “Neural networks which incorporate weather data and provide more accurate, less computationally demanding, and faster predictions of environmental event (e.g., weather) based consequences are desirable.” AND [Watt, [0005]] “The trained models are utilized to generate forecast predictions in view of predicted or scenario-based future environmental conditions.” AND [Watt, [0009]] “The trained models can include a plurality of models, and in an embodiment, three different models (Models A, B, and C) are proposed that are adapted and trained differently…Model A can be trained using the raw time-series data, while Models B and C can be trained using the de-trended data, with Model B adapted for a first environmental condition (e.g., weather), and Model C adapted for a second environmental condition (e.g., climate).” AND [Watt, [0022]] “Training a neural network in accordance with the embodiments set out herein may quantify the macro and micro relationships among weather, extreme weather, climate and climate change and features in a geo-spatial and spatio-temporal context.”, wherein the examiner interprets Watt’s plurality of differently trained machine learning models adapted for environmental conditions as well as in “geo-spatial” contexts to be the same as using a plurality of machine learning models for prediction of environmental/weather-related conditions.)
Mukherjee and Watt do not teach selecting, by the computer, a subset of the one or more actionable outputs whose correlations satisfy a threshold value; and executing, by the computer, the subset of executable actions associated with the respective event type prior to occurrence of the event occurring in the geospatial region;.
Appel teaches selecting, by the computer, a subset of the one or more actionable outputs whose correlations satisfy a threshold value; ([Appel, [0035]] “The score may be an output of an algorithm based on the result of training the cognitive model and the network 315…The vulnerability score may represent the propensity of an entity to have a vulnerability or a failure in the future…The cognitive model 315 may compute the vulnerability score of each entity of that supply chain for a given period in the future. In one embodiment, the vulnerability score may take a value in the range of 0 to 1 for each time period, where 0 represents no risk of failure, and 1 represents the highest level of certainty that a failure is likely to occur.” AND [Appel, [0036]] “In some cases, the supply chain vulnerability system 300 may determine whether the vulnerability score for a particular entity is above (or below) a threshold, and if so, recommend ways to mitigate the impact of an external event. Such threshold may be configurable and tunable for specific time and situation, i.e. the threshold is customer defined and adjustable. For example, if the threshold is set to 0.8, a mitigation strategy would be generated for one or more combination of time periods, entities, and potentially disruptive events if the vulnerability score associated with the entity during that time period and event is greater than 0.8.”, wherein the examiner interprets generating mitigation strategies only when a future-event vulnerability score exceeds a configurable threshold to be the same as “selecting, by the computer, a subset of the one or more actionable outputs whose correlations satisfy a threshold value” because both are directed to selecting or generating action-oriented outputs only when a predictive score, relationship, or correlation satisfies a threshold condition.)
High teaches executing, by the computer, the subset of executable actions associated with the respective event type prior to occurrence of the event occurring in the geospatial region; ([High, [0005] “matching the detected upcoming event to one or more categories…retrieving, from a database, static data corresponding to each of the one or more categories that matches the detected upcoming event” AND [High, [0006]] “The one or more memory devices store executable code effective to cause the one or more processors to monitor a plurality of real-time data sources to detect an upcoming event, match the detected event to one or more categories, and retrieve, from a database, static data corresponding to each of the one or more categories that matches the detected upcoming event.”, wherein the examiner interprets executable code effective to cause one or more processors to monitor real-time data sources, detect an upcoming event, match the detected event to one or more categories, and retrieve database information corresponding to the matched categories to be the same as “executing, by the computer, the subset of executable actions associated with the respective event type prior to occurrence of the event occurring in the geospatial region” because both are directed to computer-executed actions performed in response to an upcoming event and associated event type before the upcoming event occurs.)
Mukherjee, Watt, Appel, High, and the instant application are analogous art because they are all directed to computer-implemented methods for utilizing predictions of future real-world events or environmental conditions with contextual information to generate, select, or execute actionable outputs before the predicted event occurs.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the weather intelligence method disclosed by Mukherjee to include the “training a plurality of different machine learning models” disclosed by Watt. One would be motivated to do so to efficiently improve the accuracy, speed, and environmental-condition specificity of Mukherjee’s weather prediction system, as suggested by Watt ([Watt, [0005]], “more accurate, less computationally demanding, and faster predictions of environmental event (e.g., weather) based consequences”).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to include the “determine whether the vulnerability score for a particular entity is above (or below) a threshold, and if so, recommend ways to mitigate the impact of an external event” disclosed by Appel. One would be motivated to do so to effectively select only actionable outputs that satisfy a sufficient threshold condition indicating relevance or expected impact, as suggested by Appel ([Appel, [0036]], “recommend ways to mitigate the impact of an external event”).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to include the “The one or more memory devices store executable code effective to cause the one or more processors to monitor a plurality of real-time data sources to detect an upcoming event, match the detected event to one or more categories, and retrieve, from a database, static data corresponding to each of the one or more categories that matches the detected upcoming event.” disclosed by High. One would be motivated to do so to efficiently execute concrete computer-directed actions before a future event occurs once detected and matched, as suggested by High ([[High, [0005] “matching the detected upcoming event to one or more categories…retrieving, from a database, static data corresponding to each of the one or more categories that matches the detected upcoming event” AND [High, [0006]] “The one or more memory devices store executable code effective to cause the one or more processors to monitor a plurality of real-time data sources to detect an upcoming event, match the detected event to one or more categories, and retrieve, from a database, static data corresponding to each of the one or more categories that matches the detected upcoming event.”). Claims 14 and 20 are analogous to claim 1, aside from minute differences, but the same rejection in content is analogous and can apply to all three.
Regarding claim 3, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Mukherjee further teaches:
wherein correlating, by the computer, the one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region is based on a predetermined mapping relationship between the plurality of prediction results and the semantic information. ([Mukherjee, [0043]] “Upon classifying each of the time intervals from the weather forecast data for a given geographic region, the weather classifier 215 updates a region context 212 for that geographic region with the classifications. These classifications can also feed back into the modeling component 210 in order to analyze correlations and determine thresholds for unusual temperatures, such as the gradients that may indicate a time interval is unusually hot or cold.” AND [Mukherjee, [0044]] “At a regular interval of time (e.g., once every 3 hours), the publisher 230 queries the region contexts 212 and artifact table 222 to determine which consumer processes 250 to send which lists of geographic regions. The publisher 230 extracts the artifact data of each registered service by the right namespaces and the right consumers and uses that information to identify the consumer processes 250 to receive lists of geographic regions experiencing an unusual weather condition for a given time period. For example, the publisher 230 can create three lists from the region context 212 data: a list of geographic regions determined to be unusually hot, a list of geographic regions determined to be unusually cold, and a list of geographic regions determined to be inclement. For each consumer process 250 registered to receive lists of unusually hot regions, the publisher 230 sends the unusually hot region list to that consumer process 250.” AND [Mukherjee, [0036]] “A parser 220 on the service provider side parses the handshake data, extracts artifact details identifying the type of weather trigger and the consumer process 250, and saves the details in an artifact table 222 for use by a publisher process 230 after processing weather forecast data.”, wherein the examiner interprets classifying time intervals for a given geographic region, updating a region context with the classifications, querying the region contexts and artifact table to determine which consumer processes receive which lists of geographic regions, creating lists of geographic regions determined to be unusually hot, unusually cold, or inclement, and using artifact details identifying the type of weather trigger and consumer process to be the same as “wherein correlating, by the computer, the one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region is based on a predetermined mapping relationship between the plurality of prediction results and the semantic information” because Mukherjee uses stored artifact/registration information to determine how predicted weather-event classifications for geographic regions are associated with registered consumer-process information. The examiner further interprets the lists of geographic regions determined to be unusually hot, unusually cold, or inclement as the claimed plurality of prediction results of the respective event type, the geographic regions as the claimed geospatial region, and the artifact data/registered service/consumer-process information as the claimed semantic information because the stored artifact data defines which predicted event-type/geographic-region outputs are associated with which registered consumer process.)
Regarding claim 4, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Appel teaches:
wherein generating, by the computer for the user device, one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information comprises: providing the one or more prediction results and the semantic information to a trained machine learning model; and receiving, from the trained machine learning model, an output including data that identifies the one or more actionable outputs. ([Appel, [0031]] “Examples of data that may be collected include: weather/climate data (like temperature, humidity, wind, precipitation & lightening forecast), natural disaster data statistical, remote sensing imagery, price and other market data & historical risk data, supply chain network data about the entity information (e.g., from official news website, social media, etc.) Each type of data may be processed according to a unique data structure for that type of data. For instance, text data may be processed with Natural Language Processing (NLP) methods or text mining algorithms for topic detection; weather, climate and remote sensing data may be processed with geo-referential tools and databases then fed into physical/statistical models.” AND [Appel, [0034]] “This cognitive model 315 may be implemented as any supervised machine learning model that can use the network structure over time. One example of a cognitive model 315 would be a neural network… Once the model is trained, the network 315 may facilitate predictions for the supply chain based on the aggregated historical data.” AND [Appel, [0035]] “Once the cognitive model 315 is trained, the supply chain vulnerability system 300 may compute a vulnerability score for each entity. The score may be an output of an algorithm based on the result of training the cognitive model and the network 315.” AND [Appel, [0036]] “In some cases, the supply chain vulnerability system 300 may make one or more recommendations for a mitigation strategy. For example, the supply chain vulnerability system 300 may determine whether the vulnerability score for a particular entity is above (or below) a threshold, and if so, recommend ways to mitigate the impact of an external event.”, wherein the examiner interprets Appel’s weather/climate data, market data, historical risk data, and supply chain network data about entity information to be the same as the one or more prediction results and the semantic information because they are prediction-related and contextual data used to determine vulnerability and responsive recommendations. The examiner further interprets Appel’s cognitive model implemented as a supervised machine learning model or neural network to be the same as the trained machine learning model. The examiner further interprets Appel’s vulnerability score output from the trained cognitive model to be the same as an output including data that identifies the one or more actionable outputs because the vulnerability score identifies whether a particular entity has sufficient future vulnerability to trigger a mitigation recommendation. The examiner further interprets Appel’s recommendation to “mitigate the impact of an external event” to be the same as the one or more actionable outputs because both are directed to action-oriented outputs generated in response to predicted future external events.)
Mukherjee, Watt, Appel, High, and the instant application are analogous art because they are all directed to computer-implemented methods for utilizing predictions of future real-world events or environmental conditions with contextual information to generate, select, or execute actionable outputs before the predicted event occurs.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to further modify the computer-implemented method of claim 1 disclosed by Mukherjee, Watt, Appel, and High to include the “cognitive model 315 may be implemented as any supervised machine learning model…score may be an output of an algorithm based on the result of training the cognitive model and the network 315” disclosed by Appel. One would be motivated to do so to effectively use a trained machine learning model to generate output data identifying whether mitigation/action recommendations should be generated for predicted future external events, as suggested by Appel ([Appel, [0036]], “recommend ways to mitigate the impact of an external event”).
Thus, claim 4 is non-patent eligible. Claim 17 is analogous to claim 4, aside from claim type, and thus the same rejection will apply.
Regarding claim 5, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Mukherjee further teaches:
wherein the one or more actionable outputs is automatically executed by the user device. ([Mukherjee, page 1] “can then programmatically enable a trigger to transmit a service-related offer associated with the weather anomaly to user devices located within one of the geographic regions where that weather anomaly is determined.” and [Mukherjee, [0056]] “offers can cause a user device to display the offer on a user interface of an application (e.g., through the user of a push notification)”, wherein the examiner interprets “enable a trigger to transmit a service-related offer” to be the same as the one or more actionable outputs because they are both directed to an output generated for a user in response to a predicted real-world event and provided to the user device for action. The examiner further interprets “offer” to be the same as the one or more actionable outputs because they are both directed to an output delivered to a user device for the user to act upon, and wherein the examiner interprets “cause a user device to display” to be the same as is automatically executed by the user device because they are both directed to the user device automatically performing an action using the provided output without requiring the user to manually execute the output.) Thus, claim 5 is non-patent eligible. Claim 18 is analogous to claim 5, aside from claim type, and thus the same rejection will apply.
Regarding claim 6, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Appel further teaches wherein the one or more actionable outputs ([Appel, [0036]] “In some cases, the supply chain vulnerability system 300 may make one or more recommendations for a mitigation strategy. For example, the supply chain vulnerability system 300 may determine whether the vulnerability score for a particular entity is above (or below) a threshold, and if so, recommend ways to mitigate the impact of an external event.” wherein the examiner interprets “recommendation for a mitigation strategy” to be the same as “one or more actionable outputs” because they are both describing actionable outputs generated by the system (recommendations/mitigation strategies).
Mukherjee teaches is displayed by the user device to a user using a graphic user interface (GUI). ([Mukherjee, [0056]] “In some implementations, offers can cause a user device to display the offer on a user interface of an application (e.g., through the user of a push notification) registered with the platform environment.”, wherein the examiner interprets “user interface” of an application to be the same as “the one or more actionable outputs is displayed by the user device to a user using a graphic user interface (GUI)” because they are both describing actionable outputs (offers/recommendations) being displayed to users on a user device through a user interface.)
Mukherjee, Watt, Appel, High, and the instant application are analogous art because they are all directed to generating actionable outputs in response to predicted or forecast real-world/external events, and presenting those actionable outputs to a user via a user device interface.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the computer-implemented method of claim 1 disclosed by Mukherjee, Watt, Appel, and High to include the notification system disclosed by Mukherjee. One would be motivated to do so to effectively present the mitigation recommendations/actionable outputs to users through the user device interface so that users can timely receive and act upon the recommendations, as suggested by Mukherjee ([Mukherjee, [0056]] “through the user of a push notification”). Claim 19 is analogous to claim 6, aside from claim type, and thus the same rejection will apply.
Regarding claim 7, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Mukherjee further teaches wherein the event comprises a natural event or condition. ([Mukherjee, [0014]] “The weather intelligence system determines, using the weather forecast data for each of the geographic regions, a set of geographic regions predicted to experience an unusual weather condition, or weather anomaly, during a particular time interval. The weather alert can be the result of abnormal weather conditions such as unusually hot or cold temperatures and inclement weather.”, wherein the examiner interprets “weather anomaly” and “unusual weather condition” to be the same as “a natural event or condition” because they are both identifying the type of event being detected/processed by the system as a naturally occurring phenomenon (weather conditions, temperature anomalies, precipitation) rather than a human-initiated or artificial event.)
Regarding claim 8, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 7, (see rejection for claim 7).
Mukherjee further teaches wherein the natural event or condition comprises a natural hazardous event. ([Mukherjee, [0014]] “The weather intelligence system determines, using the weather forecast data for each of the geographic regions, a set of geographic regions predicted to experience an unusual weather condition, or weather anomaly, during a particular time interval. The weather alert can be the result of abnormal weather conditions such as unusually hot or cold temperatures and inclement weather.”, wherein the examiner interprets “unusual weather condition , or weather anomaly” to be the same as the natural event or condition because they are both directed to a naturally occurring real-world condition/event, and wherein the examiner interprets “abnormal weather conditions such as unusually hot or cold temperatures and inclement weather” to be the same as a natural hazardous event because they are both directed to naturally occurring adverse conditions that can create hazardous impacts.)
Regarding claim 9, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Mukherjee further teaches wherein the semantic information comprises information corresponding to a product, a service, and/or a provider thereof. ([Mukherjee, [0045]] “A consumer interface 270 receives the appropriate list of geographic regions from the service provider and creates corresponding weather-related triggers for each of the geographic regions on the list. A rules engine 280 can then index user profiles from a database of user profiles 284 to determine which users registered with the environment are currently located within (or near) any of the geographic regions on the list. The rules engine 280 can also filter the user profiles by one or more programmed business rules 282, which enterprise customers can configure to target desired characteristics for offer recipients. For example, a retail location may configure an offer to send a push notification offering a discount on cold drinks to user devices of registered users within an unusually hot geographic region, but only for users who are not already regular customers of the business.”, wherein the examiner interprets “one or more programmed business rules 282, which enterprise customers can configure” to be the same as the semantic information comprises information corresponding to a product, a service, and/or a provider thereof because they are both directed to configurable, meaning-bearing context information supplied by an enterprise/provider that specifies what offering is being provided and to whom. The examiner further interprets “cold drinks” to be the same as a product because they are both directed to a specific item being offered, and wherein the examiner interprets “retail location” to be the same as a provider thereof because they are both directed to an entity that provides the product and/or service referenced by the offer, and wherein the examiner interprets “offer” and “push notification offering a discount” to be the same as a service because they are both directed to an offering made available to users by the provider in connection with the system’s trigger-based operation.)
Regarding claim 10, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Mukherjee further teaches:
wherein obtaining, by the computer and from the plurality of machine learning models, the plurality of prediction results in association with the event ([Mukherjee, [0014]] “The weather intelligence system determines, using the weather forecast data for each of the geographic regions, a set of geographic regions predicted to experience an unusual weather condition, or weather anomaly, during a particular time interval. The weather alert can be the result of abnormal weather conditions such as unusually hot or cold temperatures and inclement weather. The weather intelligence system determines, through a machine-learning process, whether the weather forecast data indicates conditions that people would generally consider to be unusually hot or cold.”, wherein the examiner interprets “through a machine-learning process” to be the same as obtaining, by the computer and from the machine learning model because they are both directed to a computer obtaining predictive output generated by a machine learning model. The examiner further interprets “a set of geographic regions predicted to experience” to be the same as the plurality of prediction results because they are both directed to multiple prediction outputs, and wherein the examiner interprets “weather anomaly , or unusual weather condition” to be the same as the event because they are both directed to a real-world event/condition being predicted.)
is based on the semantic information. ([Mukherjee, [0025]] “Once a geographic region identifier (e.g., a zip code) is passed in the header of the API, the weather data provider 130 returns the weather forecast data for that zip code.”, wherein the examiner interprets “a geographic region identifier ( e.g. , a zip code )” to be the same as the semantic information because they are both directed to a meaningful descriptor used to condition/select the information used for prediction. The examiner further interprets “returns the weather forecast data for that zip code” to be the same as obtaining … the plurality of prediction results … is based on the semantic information because they are both directed to the predictive results being obtained based on an input descriptor (here, the region identifier) that determines what data is used to generate the prediction results.)
Regarding claim 11, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Mukherjee further teaches:
wherein generating the one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information is in response to determining that at least one of the one or more parameters of the event satisfies a predetermined threshold. ([Mukherjee, [0051]] “For each time interval, the weather intelligence system determines whether the temperature of the time interval exceeds its moving average, combined with programmed floor and ceiling thresholds, in order to classify the time interval as unusually hot or cold (330).” AND [Mukherjee, [0032]] “Upon activation of a trigger and satisfaction of any business rules associated with the offer, the network computer system 100 may send the offer to a user device 160.” AND [Mukherjee, [0045]] “A consumer interface 270 receives the appropriate list of geographic regions from the service provider and creates corresponding weather-related triggers for each of the geographic regions on the list…For example, a retail location may configure an offer to send a push notification offering a discount on cold drinks to user devices of registered users within an unusually hot geographic region, but only for users who are not already regular customers of the business.”, wherein the examiner interprets “temperature of the time interval” to be the same as “at least one of the one or more parameters of the event” because temperature is a quantitative parameter of the predicted weather event. The examiner further interprets “programmed floor and ceiling thresholds” to be the same as “a predetermined threshold” because both are predefined threshold values used to evaluate whether the event parameter satisfies a threshold condition. The examiner further interprets classifying the time interval as unusually hot or cold, creating corresponding weather-related triggers for geographic regions, and sending an offer to a user device upon activation of a trigger and satisfaction of business rules to be the same as “generating the one or more actionable outputs based on the correlations between the one or more prediction results and the semantic information” “in response to determining that at least one of the one or more parameters of the event satisfies a predetermined threshold” because Mukherjee generates/sends the offer after the weather event parameter satisfies the programmed threshold condition and after the predicted weather/geographic-region information is associated with business-rule/user-context information.)
Regarding claim 13, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Appel further teaches:
wherein correlating, by the computer, the one or more prediction results of the respective event type of the plurality of prediction results to the semantic information for the geospatial region is by using a machine learning model trained on historical events and historical data in correspondence to the semantic information. ([Appel, [0015]] “Historical data such as weather data, climate data, news, social media, market data and data related to the relationships among supply chain entities may be aggregated in a dependency network and used to train a cognitive model.” AND [Appel, [0031]] “Examples of data that may be collected include: weather/climate data (like temperature, humidity, wind, precipitation & lightening forecast), natural disaster data statistical, remote sensing imagery, price and other market data & historical risk data, supply chain network data about the entity information” AND [Appel, [0034]] “This cognitive model 315 may be implemented as any supervised machine learning model that can use the network structure over time. One example of a cognitive model 315 would be a neural network.” AND [Appel, [0034]] “The past snapshots of the supply chain, (i.e., the historical data), are used to train the cognitive model by applying past known causes and tuning the (also known) effects. Once the model is trained, the network 315 may facilitate predictions for the supply chain based on the aggregated historical data.”, wherein the examiner interprets “cognitive model 315” implemented as “any supervised machine learning model” or “a neural network” to be the same as a machine learning model. The examiner further interprets “historical data such as weather data, climate data, news, social media, market data and data related to the relationships among supply chain entities” and “past snapshots of the supply chain, (i.e., the historical data)” to be the same as historical data in correspondence to the semantic information because they are directed to historical contextual/entity/event-related data used by the trained model. The examiner further interprets “applying past known causes and tuning the (also known) effects” to be the same as historical events and historical data in correspondence to the semantic information because both are directed to training the model using prior event/cause information and corresponding known results/effects. The examiner further interprets the trained cognitive model facilitating predictions based on aggregated historical data to teach using the machine learning model to correlate prediction-related information with contextual/semantic information because the trained model uses historical event, entity, weather/climate, market, and relationship data to generate future prediction outputs.)
Mukherjee, Watt, Appel, High, and the instant application are analogous art because they are all directed to computer-implemented methods for utilizing predictions of future real-world events or environmental conditions with contextual information to generate, select, or execute actionable outputs before the predicted event occurs.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the computer-implemented method of claim 1 disclosed by Mukherjee, Watt, Appel, and High to include the “past snapshots of the supply chain, (i.e., the historical data), are used to train the cognitive model by applying past known causes and tuning the (also known) effects” disclosed by Appel. One would be motivated to do so to effectively use historical event/cause information and corresponding historical data to improve model-based prediction and correlation of future event-related information with contextual information, as suggested by Appel ([Appel, [0034]], “Once the model is trained, the network 315 may facilitate predictions for the supply chain based on the aggregated historical data.”).
Claims 2 and 15 is rejected under 35 U.S.C. 103 as being unpatentable over Mukherjee in view of Watt in view of Appel in view of High further in view of US 20170169446 A1, by Li et. al. (referred herein as Li).
Regarding claim 2, Mukherjee, Watt, Appel, and High teaches The computer-implemented method of claim 1, (see rejection for claim 1).
Mukherjee further teaches:
wherein each of the plurality of prediction results comprises one or more parameters of: ([Mukherjee, page 1] “A weather intelligence system retrieves weather forecast data for a number of geographic regions. The weather intelligence system determines, using the weather forecast data for each of the geographic regions, a set of geographic regions predicted to experience a weather anomaly, or unusual weather condition, during a particular time interval.”, wherein the examiner interprets “weather forecast data for a number of geographic regions” and “during a particular time interval” to be the same as each of the plurality of prediction results comprises one or more parameters of: because they are both directed to prediction results that include parameter values describing a predicted event and at least where and when it is expected to occur.)
(1) the event; (2) a time window that the event occurs; (3) a geospatial area that the event occurs; ([Mukherjee,[0014]] “The weather intelligence system determines, using the weather forecast data for each of the geographic regions, a set of geographic regions predicted to experience an unusual weather condition, or weather anomaly, during a particular time interval.” wherein the examiner interprets “weather anomaly , or unusual weather condition”, “during a particular time interval”, and “geographic regions” to be the same as the event, “time window”, and geospatial area because they are both directed to the real-world event/condition being predicted for a particular time and place.)
(4) an intensity of the event; and ([Mukherjee, [0051]] “For each time interval, the weather intelligence system determines whether the temperature of the time interval exceeds its moving average, combined with programmed floor and ceiling thresholds, in order to classify the time interval as unusually hot or cold (330).” wherein the examiner interprets forecast “temperature” and “programmed floor and ceiling thresholds” to be the same as an intensity of the event because they are both directed to a quantitative magnitude/level used to characterize how severe or strong the predicted event/condition is.)
Mukherjee, Watt, Appel, and High does not teach (5) a probability of occurrence of the event corresponding to one or more of (2), (3), and (4).
Li teaches, (5) a probability of occurrence of the event corresponding to one or more of (2), (3), and (4). ([Li, [0027]] “the confidence factor threshold can be 80%, such that a model being considered would be selected to forecast a future demand of the product when the model achieves a confidence factor based on historic forecasting that is 80% or greater.”, [Li, [0013]] “The selected model can be applied in generating a forecasted future demand of the first product at the shopping facility over a fixed future period of time.”, and [Li, [0051]] “The forecasted results may define a forecasted demand over one or more weeks at a shopping facility for the product of interest. For example, in some implementations, the results provide a forecast for 5, 17, 25 or more weeks.”, wherein the examiner interprets “confidence factor threshold can be 80%” to be the same as a probability of occurrence of the event because they are both directed to a numeric likelihood measure associated with a predicted outcome. The examiner further interprets “over a fixed future period of time” to be the same as corresponding to one or more of (2) because they are both directed to tying the prediction (and its likelihood/confidence) to a defined time window, and wherein the examiner interprets “at a shopping facility” to be the same as corresponding to one or more of (3) because they are both directed to tying the prediction (and its likelihood/confidence) to a defined location/geospatial area.)
Mukherjee, Watt, Appel, High, Li, and the instant application are analogous art because they are all directed to prediction results for a real-world event.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the computer-implemented method claim 1 disclosed by Mukherjee, Watt, Appel, and High to include the confidence factor disclosed by Li. One would be motivated to do so to reliably include a numeric likelihood/confidence parameter with the prediction results so that downstream actions are based on sufficiently reliable prediction results, as suggested by Li ([Li, [0027]] “The confidence threshold can be set to avoid making changes to inventory that are not expected to have significant benefit.”). Claim 15 is analogous to claim 2, aside from claim type, and thus the same rejection can apply to both as above.
Conclusion
THIS ACTION IS MADE FINAL. 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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DEVAN KAPOOR whose telephone number is (703)756-1434. The examiner can normally be reached Monday - Friday: 9:00AM - 5:00 PM EST (times may vary).
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, David Yi can be reached at (571) 270-7519. 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.
/DEVAN KAPOOR/Examiner, Art Unit 2126
/DAVID YI/Supervisory Patent Examiner, Art Unit 2126
Prosecution Insights