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 .
Response to Amendment
The amendments filed 10/08/2025 have been entered. Claims 1-20 remain pending in the application.
Applicant’s amendments and arguments, with respect to claim rejections of claims 1-20 under 35 U.S.C 103 filed 07/08/2025 have been considered and are not persuasive. Therefore, the previous rejections as set forth in the previous office action will be maintained.
The applicant argues that claim the 103 rejections of independent claim 17 is deficient because the Office Action improperly relies on Polleri without establishing that Polleri is entitled to its claimed provisional priority date. Specifically, applicant contends that the Polleri Provisional does not disclose the subject matter relied upon in the Office Action – namely paragraph 50, 54, and 176. Because the examiner relies exclusively on these paragraphs to supply key limitations, applicant asserts that Polleri cannot be used as prior art for those teachings, and therefore the Office Action fails to establish a prima facie case of obviousness.
Even assuming that Polleri were entitled to the provisional date, the cited paragraphs still fail to discloses or render obvious the claimed features. The applicant asserts that Polleri does not discloses performing actions based on one or more instances of user interface input nor using a tailored machine-learning model to determine whether action conditions are satisfied in order to automatically perform computer actions as recited ion claim 17. The cited paragraphs of Polleri fail to render obvious “based on the one or more instances of user interface input being from the user or an additional user of the organization, and based on the machine-learning based condition being included in the defined one or more action conditions”
With regard to claim 1 and 20, these claims have been amended without conceding the propriety of the prior art rejections. Applicant argues that the Office Actions improperly relies on Abbondanzio to discloses amended claim features relating to receiving user interface input and confirming assignment of a machine-learning-based condition to computer actions. Applicant asserts that Abbondanzio merely discloses select an action from a displayed list, which does not satisfy the claimed sequence of receiving initial input, rendering a machine-learning-based condition, receiving confirmatory input and assigning that condition as an action condition.
The examiner respectfully disagrees.
With regard to claim 17, applicant’s argument is not persuasive because Polleri’s provisional 62/900,537 sufficiently describes the subject matter relied upon in paragraph 50. Specifically, the provisional discloses receiving user input through conversational, textual, graphical interface to define a machine learning solution and configure how predictions are generated and consumed, including identifying desired predictions, data locations, and performance metrics (provisional ¶ 6 “In one aspect, techniques can be used for defining a machine learning solution, including receiving a first input (e.g., aural, textual, or GUI) describing a problem for the machine learning solution”, and provisional ¶ 10 “A user (e.g., an application developer) can automatically configure a machine learning infrastructure via a conversational interface (e.g., a chatbot). The user can define how the machine language predictions can be consumed (e.g., “via a REST API” or “saved to file”). The user can further define the location of the data. The user can also identify what additional 20 25 services can be required (e.g., monitoring, logging, and alerting) for the machine learning infrastructure. Constraints (e.g., resources, location, security, privacy) can be identified by the user.”) The provisional further explains that end user can interact with the system through interfaces such as chatbots to control and configure aspect of the machine-learning system (provisional ¶ 39 “In some embodiments, end users can be controlling the development. If a simple interface, e.g., a drop-down menu is used, a user could interact with the chatbot to change default views of the drop-down list. The chatbot could ask what is desired in the drop-down list. The chatbot can modify the drop-down list based on user preferences. In certain embodiments, an 15 20 25 end user can interact with the model composition engine 132. In that way, an end-user can have a very bespoke interaction with the artificial intelligence system.”) These disclosures collectively support Polleri’s paragraph 50, which recites that a user interacts with an interface to identify a desired machine learning prediction and performance metrics for the machine learning model. A person ordinary skilled in the art would recognize that paragraph 50 as a predictable and routine elaboration of the provisional’s disclosure of user-driven configuration of machine-learning solutions via user interfaces, rather than the introduction of new subject matter. Accordingly, Polleri is entitled to the benefit of its provisional filing date with respect to features relied upon in paragraph 50.
Polleri’s provisional 62/900,537 sufficiently describes the subject matter relied upon in paragraph 54. Specifically, the provisional discloses monitoring operation of a machine-learning application according to KPI/QoS metrics to assure that the application is performing according to requirements and to support testing and evaluation of new or evolving applications (provisional ¶ 4 “A monitoring engine 156 monitors operation of application according to the KPI/QoS metrics 160 to assure the ML application is performing according to requirements to seamlessly test an end-to-end a new or evolving application at different scales, settings, loading, settings, 20 25 30 etc.”) The provisional further explains that machine-learning algorithms analyze performance and conduct comparative analysis of results to determine performance trends (provisional ¶ 19 “The problem could be defined as determining employee trends. A machine learning algorithm could be used to analyze the employee records, identify information regarding the performance of each employee, and conduct comparative analysis of historical reviews to determine performance trend of each of the employees”), which corresponds to comparing model execution results against performance characteristics. Additionally, the provisional discloses using test data and model outputs prior to deployment to evaluate whether the machine-learning model meets constraints, and performs as intended (provisional ¶ 37 “In various embodiments, a user can identify test data to determine the output of the machine learning model. Prior to deployment, the intelligent assistant can display optimal solutions that meet the constraints to the user.”), which aligns with using ground-truth data to test the model. A person ordinary skilled in the art would understand paragraph 54 as a predictable and routine implementation of monitoring, evaluating, and comparing machine-learning results against performance metrics as disclosed in the provisional rather than the introduction of new subject matter. Accordingly, Polleri is entitled to the benefit of its provisional filing date with respect to features relied upon in paragraph 54.
Polleri’s provisional 62/900,537 sufficiently describes the subject matter relied upon in paragraph 176. Specifically, the provisional discloses embodiment in which a state machine includes user-defined states corresponding to end-user intents and actions to be taken in those states, and further explained that, based on user input, the system determine the end’s user intent in order to determine the appropriate next action to take (provisional ¶ 16 “In some embodiments, a state machine that includes user defined states (e.g., end user intents) and actions to take in the states or from state to state may be used to implement the bot ... For example, at each state, based on the user input, the bot may determine the end user's intent in order to determine the appropriate next action to take.”). Meanwhile, paragraph 176 explains that after determining an end-user intent based on user input, the determined intent an associated parameters are provided to an action engine to determine an action to perform. A person ordinary skilled in the art would understand paragraph 176 as a predictable and routine implementation of intent determination followed by action selection as disclosed in the provisional rather than the introduction of new subject matter. Accordingly, Polleri is entitled to the benefit of its provisional filing date with respect to features relied upon in paragraph 176.
With regard to the argument that the cited references by Polleri fail to discloses “based on the one or more instances of user interface input being from the user or an additional user of the organization, and based on the machine-learning based condition being included in the defined one or more action conditions”, the examiner respectfully disagree. As recited in paragraph 50 “A model composition engine 132 can be executed on one or more computing systems (e.g., infrastructure 128). The model composition engine 132 can receive inputs from a user 116 through an interface 104. The interface 104 can include various graphical user interfaces with various menus and user selectable elements. The interface 104 can include a chatbot (e.g., a text based or voice based interface). The user 116 can interact with the interface 104 to identify one or more of: a location of data, a desired prediction of machine learning application, and various performance metrics for the machine learning model.”, Polleri discloses a user interacts with an interface to identify a desired machine-learning prediction, relevant performance metrics, and other parameters that govern operation of the machine-learning system. Under the broadest reasonable interpretation, this disclosure teaches system behavior that is based on one or more instances of user interface input, because the identification of the desired prediction and performance metrics is provided via user interaction with the interface. Polleri further teaches the user-identified desired prediction and performance metrics define how the machine-learning model operates and evaluated, which constitutes a machine learning based condition used by the system. Polleri also discloses at paragraph 176 “After the end user intent is determined based on the content by message processor 550, the determined intent (and the parameters associated with the intent) may be sent to an action engine 560. Action engine 560 may be used to determine an action to perform based on the intent” Thus, Polleri teaches that machine learning outputs are evaluated by an action engine to determine an action, which corresponds to the machine-learning based condition being included in the defined one or more action conditions, as claimed. The claim does not require an explicit recitation of a separately labeled “action condition”, but rather requires that whether an action is performed is determined based on user interface input in combination with a machine-learning based condition. Polleri’s paragraph 50 teaches precisely this by disclosing that system decisions and behavior are driven by user-provided inputs in conjunction with machine learning outputs, metrics, and action perform based on the intent, thereby meeting the claimed limitation.
With regard to claim 1 and 20, the claims disclose the one or more instances of user interface input that is provided prior to the automation interface comprising of further user interface input that confirm assignment of the machine-learning based condition to one or more computer actions. Under the broadest reasonable interpretation and as understood by one of ordinary skilled in the art, the claim requires a temporal ordering of steps, and a different type of user interaction input, wherein Abbondanzio suggests this at paragraph 14 “The mobile computing device will have, or communicate with, a display screen for displaying information about the condition and information about actions that may be taken responsive to the condition. Furthermore, the mobile computing device will have some type of input device or interface for receiving user-input, such as selecting an action from among a plurality of displayed actions” and paragraph 16 “Embodiments of the method display actions that are predetermined to be responsive to the condition. Such an action may be predetermined to be responsive to a condition though a stored association of the action to the condition, wherein the stored association may be manually input during setup of a system or may be the result of a historical record gathered over time through monitoring the actions that are user-selected to respond to any given condition and correlating the actions with the conditions.” Abbondanzio discloses a workflow in which user interface input is first received during setup to define and associate actions with condition (as recited in paragraph 16 above), followed by later user interface to confirm or invoke actions responsive to a condition (as recited in paragraph 14 above). Accordingly, Abbondanzio discloses two different user input with an order corresponds to the receiving of user interface input that confirm assignment of condition to computer actions that is provided subsequent to receiving the one or more instances of user interface input, as claimed. Although Abbondanzio describes conditions to which actions are responsive, Polleri teaches that such a condition may be a machine learning model outputs to determine conditions that corresponds to the claimed machine-learning based condition (as recited in the previous Office Action). Polleri also discloses displaying the machine learning outputs over an interface below, which corresponds to the claimed process of causing the identifier of the given machine-learning based condition to be rendered, wherein a person ordinary skilled in the art would have presented Polleri’s machine learning model output after Abbondanzio’s user action-selection/setup input so that the displayed machine learning model output that represent condition is contextually tied to the action. This ordering is a predictable and routine design choice because the user’s action selection establishes the context for which condition is relevant, and displaying the machine learning output representing condition ensures that correct action can be selected, thereby improving the overall system. The teaching of Polleri and motivation to combine the teachings are taught in the previous Office Action and are recited below.)
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.
Claim(s) 1-2, 4-8, 12, 16, 20 are being rejected under 35 U.S.C. 103 as being unpatentable over Abbondanzio et.al (US 20180167261 A1) in view of Polleri et.al (US 20210081819 A1).
Regarding claim 1,
Abbondanzio teaches the limitation “receiving one or more instances of user interface input directed to an automation interface, the one or more instances of user interface input defining one or more computer actions to be performed automatically in response to satisfaction of one or more action conditions to be defined via the automation interface” (paragraph 13 “Examples of a condition include a loss of power, component failure, high temperature, high load, or various system errors or alerts”, paragraph 16 “A plurality of actions are displayed on a screen of the mobile computing device 10, wherein the displayed actions are predetermined to be responsive to the condition. User-input is received via an interface of the mobile computing device 10, wherein the user-input selects an action from among the plurality of displayed actions, and an instruction is sent from the mobile computing device 10 to the computing system 90, wherein the instruction causes the computing system to execute the selected action”, paragraph 23 “In other embodiments, the method may further comprise automatically invoking a given action from among the plurality of actions”, and paragraph 27 “A plurality of actions are displayed on a screen of the mobile computing device 10, wherein the displayed actions are predetermined to be responsive to the condition. User-input is received via an interface of the mobile computing device 10, wherein the user-input selects an action from among the plurality of displayed actions, and an instruction is sent from the mobile computing device 10 to the computing system 90, wherein the instruction causes the computing system to execute the selected action.” Abbondanzio discloses a method, computer program product and apparatus for responding to conditions within a computing system. Within the disclosure Abbondanzio discloses a computing device comprises of an interface for user-input, wherein the user-input selects an action from among the plurality of displayed actions to be executed, wherein the action is responsive to a condition to be satisfied. An action can further be automatically invoked from a plurality of action. Examples of a condition include a loss of power, component failure, high temperature, high load, or various system errors or alerts.)
Abbondanzio teaches the limitation “identifying corresponding data associated with a plurality of past occurrences of the one or more computer actions” (paragraph 25 “The method comprises accessing, by a processor, a stored historical record of user-selected actions responsive to multiple types of conditions in a computing system, wherein each instance of a condition in the historical record is associated with a user-selected action that was taken by a user within a predetermined group of users.” Abbondanzio discloses the method comprises a stored historical record of user-selected actions responsive to multiple types of conditions in a computing system.)
Abbondanzio teaches the limitation “responsive to receiving further user interface input that is directed to the automation interface, that confirms assignment of the given machine-learning based condition to the one or more computer actions, and that is provided subsequent to receiving the one or more instances of user interface input and subsequent to causing the identifier of the given machine-learning based condition to be rendered” (paragraph 14 “The mobile computing device will have, or communicate with, a display screen for displaying information about the condition and information about actions that may be taken responsive to the condition. Furthermore, the mobile computing device will have some type of input device or interface for receiving user-input, such as selecting an action from among a plurality of displayed actions” and paragraph 16 “Embodiments of the method display actions that are predetermined to be responsive to the condition. Such an action may be predetermined to be responsive to a condition though a stored association of the action to the condition, wherein the stored association may be manually input during setup of a system or may be the result of a historical record gathered over time through monitoring the actions that are user-selected to respond to any given condition and correlating the actions with the conditions.” Abbondanzio discloses a manually user input comprising of actions that are predetermined to be responsive to a condition through an association or historical record before the user input device or interface for receiving user-input to select an action from among a plurality of displayed actions, wherein the actions is predetermined to be responsive to a condition after the user has manually input during setup of the system, which corresponds to the receiving of user interface input that confirm assignment of condition to computer actions that is provided subsequent to receiving the one or more instances of user interface input, as claimed. Although Abbondanzio describes conditions to which actions are responsive, Polleri teaches that such a condition may be a machine learning model outputs to determine conditions that corresponds to the claimed machine-learning based condition below. Polleri also discloses displaying the machine learning outputs over an interface below, which corresponds to the claimed process of causing the identifier of the given machine-learning based condition to be rendered, wherein a person ordinary skilled in the art would have presented Polleri’s machine learning model output after Abbondanzio’s user action-selection/setup input so that the displayed machine learning model output that represent condition is contextually tied to the action. This ordering is a predictable and routine design choice because the user’s action selection establishes the context for which condition is relevant, and displaying the machine learning output representing condition ensures that correct action can be selected, thereby improving the overall system. The teaching of Polleri and motivation to combine the teachings are taught below.)
Abbondanzio teaches the limitation “assigning, in one or more computer-readable media, the given machine-learning based condition as one of the action conditions” (paragraph 16 “Such an action may be predetermined to be responsive to a condition though a stored association of the action to the condition, wherein the stored association may be manually input during setup of a system or may be the result of a historical record gathered over time through monitoring the actions that are user-selected to respond to any given condition and correlating the actions with the conditions” Abbondanzio discloses an action may be predetermined to be responsive to a condition though a stored association of the action to the condition, suggesting the assigning process of a condition responsive to an action, wherein the condition may further include a machine learning model output which corresponds to a machine-learning based condition as claimed based on the combined teaching of Polleri below.)
Abbondanzio does not teach the limitation “generating, based on the corresponding data, a corresponding metric for each of a plurality of machine-learning based conditions, wherein the corresponding metrics each indicate how often a corresponding one of the plurality of machine-learning based conditions would have been considered satisfied based on the corresponding data”. However, Polleri teaches this limitation (paragraph 0009 “The machine learning techniques can monitor and evaluate the outputs of the machine learning model to allow for feedback and adjustments to the model”, paragraph 47 “A monitoring engine 156 can monitor operation of the machine learning applications 112 according to the KPI/QoS metrics 160 to assure the machine learning application 112 is performing according to requirements”, paragraph 53 “The metrics can include inference query metrics, performance metrics, sentiment metrics, and testing metrics. The metrics can be received from a user 116 through a user interface 104”, and paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model. Example training metrics can include classification accuracy, logarithmic loss, area under curve, F1 Score, mean absolute error, and mean squared error. The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made.” Polleri discloses systems and methods for an intelligent assistant that can be used to enable a user to generate a machine learning system to assist a user. Within the disclosure, Polleri discloses machine learning model that provide outputs, wherein these outputs correspond to machine learning-based conditions within the claim. Polleri discloses a monitoring engine can monitor operation of the machine learning applications by applying various metrics, wherein the metrics may be applied onto machine learning outputs as understood by a person ordinary skilled in the art, to assure the performance according to requirements. The metrics may include various training metrics, which may be input by a user, such as accuracy metric as a ratio of number of correct outputs versus the total number of outputs provided to evaluate the accuracy, which correspond to how often a corresponding one of the plurality of machine-learning based conditions would have been considered satisfied within the claim.)
Abbondanzio does not teach the limitation “causing an identifier of a given machine-learning based condition, of the plurality of machine-learning based conditions, to be rendered at the automation interface to which one or more instances of user interface input are directed”. However, Polleri teaches this limitation (paragraph 50 “The model composition engine 132 can receive inputs from a user 116 through an interface 104. The interface 104 can include various graphical user interfaces”, and paragraph 272 “A model execution system 1018 may access the trained machine-learning models 1015, provide and format input data to the trained models 1015 ... The outputs of the trained models 1015 may be provided to client devices 1050 or other output systems via the API 1012 and/or user interface components 1014. Further, the outputs of the trained models 1015 may include not only a prediction of the outcome of the code integration request (e.g., approved or denied) but also various related data such as a confidence value associated with the prediction” Polleri discloses the outputs of the trained model may be provided to the client device via the API or user interface, wherein each output being displayed on the interface corresponds to an identifier of the given machine-learning based condition within the claim. Polleri also discloses the model execution system can provide and format input data to the trained models, suggesting a communication with an interface to receive user input data, corresponding to the claimed one or more instances of user interface input.)
Abbondanzio does not teach the limitation “wherein causing the identifier of the given machine-learning based condition to be rendered is based on the corresponding metric for the given machine-learning based condition”. However, Polleri teaches this limitation (paragraph 272 “The outputs of the trained models 1015 may be provided to client devices 1050 or other output systems via the API 1012 and/or user interface components 1014. Further, the outputs of the trained models 1015 may include not only a prediction of the outcome of the code integration request (e.g., approved or denied) but also various related data such as a confidence value associated with the prediction” Polleri discloses outputs of the trained ML models is provided through the API, which may include various corresponding data such as a confidence value associated with each output, suggesting that each output may be configure by a person ordinary skilled in the art to be provided based on a satisfaction from the metric as disclosed above, wherein each output being displayed on the interface corresponds to an identifier of the given machine-learning based condition within the claim.)
Before the effective filing date, it would have been obvious to a person ordinary skilled in the art to combine the teaching of a method, computer program product and apparatus for responding to conditions within a computing system by Abbondanzio with the teaching of systems and methods for an intelligent assistant that can be used to enable a user to generate a machine learning system to assist a user by Polleri. The motivation to do so is referred to in Polleri’s disclosure (paragraph 9 “Certain aspects and features of the present disclosure relate to machine learning platform that generates a library of components to generate machine learning models and machine learning applications. The machine learning infrastructure system allows a user (i.e., a data scientist) to generate machine learning applications without having detailed knowledge of the cloud-based network infrastructure or knowledge of how to generate code for building the model”, paragraph 14 “The proposed system can use best available models at the time of construction to solve problems using the machine learning application. An adaptive pipelining composition service can identify and incorporate one or more new models into the machine learning application. The machine learning application with the new model can be tested off-line with the results being compared with ground truth data. If the machine learning application with the new model outperforms the previously used model, the machine learning application can be upgraded and auto-promoted to production”, paragraph 176 “After the end user intent is determined based on the content by message processor 550, the determined intent (and the parameters associated with the intent) may be sent to an action engine 560. Action engine 560 may be used to determine an action to perform based on the intent (and the parameters associated with the intent) and the current state (or context) of a state machine as described above.” Polleri discloses various benefits of the invention and how it may be integrated into the teaching by Abbondanzio. The teaching by Polleri provides a chatbot that may provide assistance to users in generating machine learning models and machine learning applications without having detailed knowledge of the cloud-based network infrastructure or knowledge of how to generate code for building the model. Furthermore, the system also comprises an action engine that obtain communication from a user to command the system to perform an action associated with the user. While Abbondanzio only discloses a system that provide access to historical record of user-selected actions responsive to multiple types of conditions in a computing system, but does not mention applying a machine learning model to determine the best action in responsive to a computing system condition, a person ordinary skilled in the art may combine the teaching of Abbondanzio with the teaching of the system by Polleri such that the user may interact with the bot to command various action to be performed automatically within the computing system, and the system may relies on its library data comprising of various machine learning infrastructures, metrics data as well as incorporate historical record of user-selected actions to determine an output of the best action to take in responsive to a condition of the computing system. Thus, the teaching by Polleri may be incorporated into the teaching by Abbondanzio for further improvement.)
Regarding claim 2 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “The method of claim 1, wherein the content of the identifier is based on the corresponding metric, and wherein the content comprises a visual display of the corresponding metric” (paragraph 272 “The outputs of the trained models 1015 may be provided to client devices 1050 or other output systems via the API 1012 and/or user interface components 1014. Further, the outputs of the trained models 1015 may include not only a prediction of the outcome of the code integration request (e.g., approved or denied) but also various related data such as a confidence value associated with the prediction” Polleri discloses outputs of the trained ML models is provided through the API, which may include various corresponding data such as a confidence value associated with each output, wherein each output being displayed on the interface corresponds to an identifier of the given machine-learning based condition within the claim, wherein the confidence value is a metric as disclosed above, in which the confidence value is displayed along with the corresponding output on the interface, suggesting the visual display of the corresponding metric content of the identifier within the claim.)
Regarding claim 4 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “The method of claim 1, wherein causing the identifier of the given machine-learning based condition to be rendered is based on the corresponding metric, for the given machine-learning based condition, satisfying a display threshold” (paragraph 95 “the monitoring engine can evaluate one or more QoS or KPI metrics to determine if the model meets the performance specifications ... the machine learning platform can inform the user of the monitored values, and alert the user if the QoS/KPI metrics fall outside prescribed thresholds” Polleri discloses the machine learning platform can inform the user of the monitored values such as outputs of the machine learning model, and alert the user if the metrics fall outside prescribed thresholds, suggesting that an output may be evaluated against a metric to determine if the model meets the performance specification, wherein the metric may follow a prescribed thresholds to indicate if the metric fall outside of the threshold. A person ordinary skilled in the art may configure the prescribed thresholds as a display threshold, such that the metric needs to follow and not stay out of bound of the threshold to allow the machine learning output to be displayed on the interface.)
Regarding claim 5 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “The method of claim 1, wherein further comprising: preventing any identifier of an additional machine-learning based condition, of the plurality of machine-learning based conditions, from being rendered at the automation interface, wherein the preventing is based on the corresponding metric, for the additional machine-learning based condition, failing to satisfy a display threshold” (paragraph 95 “the monitoring engine can evaluate one or more QoS or KPI metrics to determine if the model meets the performance specifications ... the machine learning platform can inform the user of the monitored values, and alert the user if the QoS/KPI metrics fall outside prescribed thresholds” Polleri discloses the machine learning platform can alert the user if the metrics fall outside prescribed thresholds, suggesting that an output may be evaluated against a metric to determine if the model meets the performance specification, wherein the metric may follow a prescribed thresholds to indicate if the metric fall outside of the threshold. A person ordinary skilled in the art may configure the prescribed thresholds as a display threshold, such that the metric needs to follow and not stay out of bound of the threshold to allow the machine learning output to be displayed on the interface.)
Regarding claim 6 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “causing, based on the corresponding metric for the given machine-learning based condition satisfying a pre-selection threshold, the identifier of the given machine-learning based condition to be pre-selected, in the automation interface, as one of the action conditions” (paragraph 96 “At 318, the functionality includes training the machine learning model with predictions judged against QoS/KPIs”, and paragraph 133 “A digital assistant 406 may use an NLP engine and/or a machine learning model (e.g., an intent classifier) to map end user utterances to specific intents (e.g., specific task/action or category of task/action that the chatbot can perform)” Polleri discloses the bot system uses machine learning model to derive task/action that can be performed, wherein the machine learning model may provide output which is evaluated against metrics, wherein a person ordinary skilled in the art can configure the output to be related to an action to be performed such that the model provide an output related to the action to be performed, in which the output is evaluated against the metric for satisfaction based on the metric threshold, and the output corresponding to the action to be performed is finally selected.)
Abbondanzio teaches the limitation “wherein the further user interface input that confirms assignment of the given machine-learning based condition to the one or more computer actions is a selection of an additional interface element that occurs without other user interface input that alters the pre-selection of the given machine-learning based condition” (paragraph 12 “the method may include storing user-selected actions responsive to multiple types of conditions in the computing system over a period of time to form the stored historical record. The stored historical may, for example, include a separate entry for each instance that an action was selected to respond to a condition. Accordingly, each entry may identify the condition, and the action selected.” Abbondanzio discloses a separate entry, wherein a person ordinary skilled in the art may configure this separate entry to be displayed as a separate element on the interface, in which the entry comprises of a computing action responsive to a condition, wherein the condition may be related to the machine learning output based on the teaching combination with Polleri, wherein this separate displayed entry does not interfere with the interface that determine action to be performed based on machine learning model outputs and metrics as disclosed above.)
Regarding claim 7 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “processing the corresponding data, using a given machine-learning model for the machine-learning based condition, to generate a plurality of corresponding values” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model. Example training metrics can include classification accuracy, logarithmic loss, area under curve, F1 Score, mean absolute error, and mean squared error. The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made”. Polleri discloses obtaining training metrics, wherein training metrics include metric such as accuracy metric as a ratio of the number of correct outputs against the total number of outputs obtained, wherein the ratio suggests the corresponding value and the prediction may be understood as the output of the machine learning model, which suggest the machine-learning based condition.)
Polleri teaches the limitation “generating the metric based on the plurality of corresponding values” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model. Example training metrics can include classification accuracy, logarithmic loss, area under curve, F1 Score, mean absolute error, and mean squared error. The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made”. Polleri discloses the ratio of the number of correct outputs against the number of outputs obtained, wherein the prediction may be understood as the output of the machine learning model, such that the ratio is considered as an accuracy metric.)
Regarding claim 8 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “The method of claim 7, wherein the plurality of corresponding values are probabilities, and wherein generating the metric comprises generating the metric as a function of the probabilities” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model. Example training metrics can include classification accuracy, logarithmic loss, area under curve, F1 Score, mean absolute error, and mean squared error. The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made”. Polleri discloses the ratio of the number of correct outputs against the total number of outputs obtained, wherein the prediction may be understood as the output of the machine learning model, wherein this ratio suggests probabilities as understood by a person ordinary skilled in the art. Thus, the accuracy metric is a function of probabilities based on the computation to obtain the ratio.)
Regarding claim 12 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “The method of claim 1, wherein identifying the corresponding data comprises identifying the corresponding data based on it being for a user that provided the user interface input, or being for an organization of which the user is a verified member” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model. Example training metrics can include classification accuracy, .... The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made.” And paragraph 130 “A text utterance, input by the user 408 or generated from converting speech input to text form, can be a text fragment, a sentence, multiple sentences, and the like. Digital assistant 406 is configured to apply natural language understanding (NLU) techniques to the text utterance to understand the meaning of the user input” Polleri discloses the system comprises of digital assistant to apply natural language understanding (NLU) techniques to the text utterance to understand the meaning of the user input, wherein the user input may be a training metric, suggesting that the system utilize natural language understanding (NLU) techniques to identify user input as the user interact with the user interface of the overall system.)
Regarding claim 16 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches the limitation “The method of claim 1, wherein identifying the corresponding data associated with the plurality of past occurrences of the one or more computer actions, and generating the corresponding metrics based on the corresponding data, both occur prior to receiving the one or more instance of user interface input” (paragraph 73 “The third user input can include training metrics. The training metrics help evaluate the performance of the model”, paragraph 270 “other components within the environment (e.g., historical data stores ...) may interface with one or more application programming interfaces (APIs) 1012 and/or user interface components 1014 supported by the prediction server 1010, to train and generate machine learning models for predicting outcomes” and paragraph 221 “In various embodiments, the technique can include a fourth input. The fourth input can include additional services required for the machine learning model”. Polleri discloses the machine learning model to train and generate machine learning models utilizing historical data for generating outcomes, wherein the outcome may be evaluated against a third input of training metrics to determine if the output of the trained machine learning model corresponding with an action of the bot system is satisfied with regard to the metric. Polleri further discloses a fourth input of additional services required for the machine learning model, suggesting that the process to generate the output of the machine learning model trained based on historical data and evaluated against the metric occurs before the fourth input, wherein the fourth input suggest the one or more instance of user interface input within the claim.)
Regarding claim 20, the applicant is directed to the rejection of claim 1 above, because the claim recites similar limitations, thus they are rejected under the same rationale.
Claim(s) 3, 13-15 are being rejected under 35 U.S.C. 103 as being unpatentable over Abbondanzio et.al (US 20180167261 A1) in view of Polleri et.al (US 20210081819 A1), further in view of Phillipps et.al (US 20140358825 A1).
Regarding claim 3 depends on claim 1, thus the rejection of claim 1 is incorporated.
Abbondanzio/Polleri does not teach the limitation “The method of claim 1, wherein the display characteristics of the identifier are based on the corresponding metric, and wherein the display characteristics comprise a size of the identifier and/or a position of the identifier in the automation interface”. However, Phillipps teaches this limitation (paragraph 56 “The one or more results modules 102 may allow a user to move the result objects around the graph, re-size the objects by stretching or pinching, or otherwise manipulate or adjust the objects. For example, in certain embodiments, an x-axis may represent time; a y-axis may represent cost, investment, man-hours, resources, or the like; a size of an object may represent a size or reach of results or targets”, and paragraph 58 “The one or more results modules 102 may use a movement, re-sizing, or other adjustment of the circles or other objects on the graph is an input to a results data structure to update the positions and sizes of other objects on the graph” Phillipps discloses an accessible user interface for machine learning results. Within the disclosure, Phillipps discloses the result module displaying a dynamic graph with multiple circles or other objects which represents an action or result of a machine learning model. The module allows a user to move, manipulate or adjust the result objects around the graph to update the positions and sizes of the objects, thus suggest that the user interface by Phillipps displays the position and size of one or more object, wherein the one or more objects represent the one or more machine learning results, wherein the machine learning results may be the same as the machine learning model outputs that suggest the identifier within the claim as disclosed by Polleri based on the teaching combination below.)
Before the effective filing date, it would have been obvious to a person ordinary skilled in the art to combine the teaching of a method, computer program product and apparatus for responding to conditions within a computing system by Abbondanzio, and the teaching of systems and methods for an intelligent assistant that can be used to enable a user to generate a machine learning system to assist a user by Polleri with the teaching of an accessible user interface for machine learning results by Phillipps. The motivation to do so is referred to in Phillipps’s disclosure (paragraph 51 “By using historical customer data sets to understand the interaction between the various actions, the one or more results modules 102 may achieve a level of confidence in using the historical data to generate patterns to predict future events, recommended actions, or the like using machine learning”, paragraph 60 “A user, in one embodiment, may directly manipulate a graphical object representing a goal, for example a familiar business graph such as a line chart, and see a graphical representation of the actions predicted to have the greatest effect on achieving the goal. The results module 102 may present such controls graphically or in another manner”, and paragraph 98 “For example, the display module 204 may provide a scrubbing slider which a user may drag to adjust a time of an animated display of machine learning results, to re-watch or further interact with the visualization and/or previous user input to machine learning parameters. Such tools may supplement or improve the experiential nature of the visualization provided by the display module 204.” Phillipps discloses the benefit of the invention, in which a graphical display of an interactive interface is provided for more efficient user’s interactions with regard to machine learning model. For instance, the interface provides interaction with historical data to generate patterns to predict future events, recommended actions, or the like using machine learning. The user may manipulate a graphical object representing various aspects of a learning process of the machine learning model. The interface provides tools that improve the experiential nature of the visualization of the display module. While Abbondanzio discloses various actions within a computing system responsive to one or more conditions and Polleri discloses the agent system with the interface to provide machine learning, a person ordinary skilled in the art may further incorporate the interface as disclosed by Phillipps to incorporate more advanced tools to allow better user’s interaction and configuration of machine learning model within a computing system, thus the user may obtain the result of the best action in responsive to a condition in a computing system through machine learning.)
Regarding claim 13 depends on claim 1, thus the rejection of claim 1 is incorporated.
Phillipps teaches the limitation “The method of claim 1 wherein the one or more actions comprise: modifying corresponding content, transmitting the corresponding content to one or more recipients that are in addition to the user, and/or causing a push notification of the corresponding content to be presented to the user” (paragraph 59 “The one or more results modules 102 may present a graphical user interface that allows a user or other client 104 to manipulate its content, in an interactive manner, in order to simulate pre-computed machine learning results, which may dictate a specific list of actions required to obtain a goal or desired outcome. A graphical user interface, as used herein, may include displayed visual objects as described above, a spreadsheet or table with numerical values (e.g., modifying cell contents for a particular action to see an effect on other cells, an action list, or the like).” Phillipps discloses a graphical user interface that allows a user to manipulate its content, in an interactive manner, wherein the content may be various aspects of the machine learning models such as model outputs and user-specified metrics from the teaching of Abbondanzio and Polleri.)
The motivation to combine the teaching of Abbondanzio and Polleri with the teaching of Phillipps is similar to the motivation as disclosed in claim 3 above.
Regarding claim 14 depends on claim 13, thus the rejection of claim 13 is incorporated.
Phillipps teaches the limitation “The method of claim 13, wherein the corresponding content is a corresponding electronic communication” (paragraph 71 “In the depicted embodiment, the results modules 102 are in communication over the data network 106 to display or otherwise present machine learning results to multiple users using electronic display devices 110 of different client computing devices 104.” Phillipps discloses the results modules are in communication over the data network to display machine learning results to multiple users using electronic display devices, wherein the communication over the data network suggest the corresponding electronic communication within the claim.)
Regarding claim 15 depends on claim 13, thus the rejection of claim 13 is incorporated.
Phillipps teaches the limitation “The method of claim 13, further comprising, subsequent to assigning the given machine-learning based condition as one of the action conditions: receiving given content of the corresponding content” (paragraph 46 “In certain embodiments, the results module 102 provides and/or accesses a machine learning framework allowing the results module 102 and/or clients 104 to request machine learning ensembles, to make analysis requests, and to receive machine learning results ... Machine learning, as used herein, comprises one or more modules, computer executable program code, logic hardware, and/or other entities configured to learn from or train on input data, and to apply the learning or training to provide results or analysis for subsequent data.” Phillipps discloses the result module is configured to receive machine learning results after a machine learning model performs learning and training using input data and a model’s output is provided for further analysis for subsequent data.)
Polleri teaches the limitation “determining that the given machine-learning based condition is satisfied, wherein determining that the given machine-learning based condition is satisfied comprises” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model ... The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made”, and paragraph 100 “At 326, the functionality includes monitoring predictions to evaluate whether the results are within expectations. In various embodiments, the monitoring engine can provide feedback to model execution engine to inform the user if the model is providing results within an expected range”. Polleri discloses training metrics such as accuracy metric to determine if the output from a machine learning model is accurate, suggesting the determination that the given machine-learning based condition is satisfied within the claim.)
Polleri teaches the limitation “processing features, of the given content, using a given machine-learning model for the given machine-learning based condition, to generate a value” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model ... The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made”. Polleri discloses a machine learning model to provide a machine learning output which suggest a given machine-learning model for the given machine-learning based condition within the claim, wherein the outputs may be one or more predictions, such that a ratio value is generate to indicate accuracy of the machine learning model.)
Polleri teaches the limitation “determining that the given machine-learning based condition is satisfied based on the value” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model ... The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made”. Polleri discloses a ratio of the number of correct outputs divided by the number of outputs being made, wherein a person ordinary skilled in the art can configure the ratio that represent the accuracy of the machine learning model to indicate if outputs from a machine learning model are satisfied if they have a high ratio of accuracy.)
Abbondanzio teaches the limitation “automatically performing the one or more actions based on determining that the given machine-learning based condition is satisfied” (paragraph 261 “In other embodiments, the method may further comprise automatically invoking a given action from among the plurality of actions”. Polleri discloses automatically invoking a given action from among the plurality of actions, wherein a person ordinary skilled in the art may configure the automated invocation of a given action based on the accuracy of the machine learning output as a machine learning model may be configured to train using action data and condition data based on the teaching combination of Abbondanzio with Polleri as disclosed above.)
Claim(s) 9-11 are being rejected under 35 U.S.C. 103 as being unpatentable over Abbondanzio et.al (US 20180167261 A1) in view of Polleri et.al (US 20210081819 A1), further in view of Schouten et.al (US 20190179945 A1).
Regarding claim 9 depends on claim 1, thus the rejection of claim 1 is incorporated.
Polleri teaches a part of the limitation “wherein the further user interface input that confirms assignment of the given machine-learning based condition to the one or more computer actions...” (paragraph 131 “For example, the user input 410 may request generation of a machine learning application, for example, “I want build an image classifier.” Digital assistant 406 is configured to understand the meaning of the utterance and take appropriate actions ... digital assistant 406 may then cause a machine learning application to be generated. Digital assistant 406 may end the conversation with the user by outputting information indicating that the machine learning application has been generated.” Polleri discloses the communication between a user and the bot system in an example to perform an action by a computer such as generating a machine learning application, the bot system goes through multiple process and finally provide an output confirming the action has been performed by the computer. The process to provide the machine learning output corresponding to the action to be performed suggest the assignment of the given machine-learning based condition to the one or more computer actions within the claim.)
Abbondanzio/Polleri does not teach the limitation “receiving additional user interface input that defines one or more rules-based conditions”. However, Schouten teaches this limitation (paragraph 43 “To facilitate definition of the rules 410, the records management and processing system 305 can further comprise a rules definition module 415. Generally speaking, the rules definition module 415 can comprise one or more applications executed by the records management and processing system 305 and which provide a rules definition interface 420. The rules definition interface 420 can include, for example, one or more webpages or other, similar interfaces providing elements through which an authorized user, such as an administrator or manager, can select or otherwise input conditions and corresponding actions for a new or modified rule” Schouten discloses a system that provide a method to maintain a set of rules defining conditions for processing records and associated actions to affect that processing upon satisfaction of or failure to satisfy the conditions of that rule. Within the disclosure, Schouten discloses a rules definition module to define the rule in accordance with conditions and corresponding actions.)
Abbondanzio/Polleri does not teach the limitation “..., and confirms assignment of the one or more rules-based conditions” However, Schouten teaches this limitation (paragraph 47 “The rules engine 440 can apply the rules 410 to the records 405 periodically, on demand, or upon the occurrence of predefined event or the satisfaction of one or more predefined conditions.” Schouten discloses the rules engine can apply the rules on the occurrence of the satisfaction of one or more predefined conditions, suggesting the assignment of the one or more rules based on predefined conditions, wherein this engine may be integrated into the interface system within the teaching combination of Abbondanzio/Polleri based on the teaching motivation below to allow the user interface input to display both machine learning outputs as well as rules for conditions and corresponding actions.)
Abbondanzio/Polleri does not teach the limitation “further comprising, in response to the further user interface input: assigning, in one or more computer-readable media, the one or more rules- based conditions as additional of the action conditions whose satisfaction results in automatic performance of the one or more computer actions” However, Schouten teaches this limitation (paragraph 42 “Generally speaking, a rule can comprise a definition of one or more conditions and an associated one or more actions to be performed upon satisfaction, or failure to satisfy, the conditions of that rule. Accordingly, each rule 410 maintained by the records management and processing system 305 can comprise one or more conditions for processing one or more records of the set of records 405 and at least one associated action to affect processing of the one or more records upon satisfaction of or failure to satisfy the one or more conditions of the rule 410.” Schouten discloses a rules definition interface to provide a rule, wherein the rule comprises a definition of one or more conditions and an associated one or more actions to be performed upon satisfaction of the rule.)
Before the effective filing date, it would have been obvious to a person ordinary skilled in the art to combine the teaching of a method, computer program product and apparatus for responding to conditions within a computing system by Abbondanzio, and the teaching of systems and methods for an intelligent assistant that can be used to enable a user to generate a machine learning system to assist a user by Polleri with the teaching of a system that provide a method to maintain a set of rules defining conditions for processing records and associated actions to affect that processing upon satisfaction of or failure to satisfy the conditions of that rule by Schouten. The motivation to do so is referred to in Schouten’s disclosure (paragraph 40 “Based on such predictions, selected actions can be taken on those records most in need of additional action or which will yield the greatest benefit from additional actions thereby increasing the efficiency of how and where time and resources of the records management and processing system 305 are used. To do so, the records management and processing system 305 can apply the rules to the records and assign tags to the records based on the model and the conditions defined in the applied rules. The records management and processing system 305 can then process the records according to workflows for processing the records based on the model, assigned tags, and applied rules” and paragraph 49 “therefore, the rules engine 440 can apply one or more of the rules 410 to the records saved in the database 405 to identify those records which should be further processed or given further scrutiny and mark those records with one or more tags 425. In this way, those records found to be within normal or acceptable conditions according to the applied rules need not be subjected to further scrutiny and/or processing thus saving resources such as human effort, processing overhead, etc” Schouten discloses the benefit of the invention, which provide a method to define rule in accordance with condition and action within a record database of a computing system. The method to define rule allow better action selections within an applied system and increasing the efficiency of how and where time and resources of such system, saving resources such as human effort and processing overhead. While the action within the disclosure by Schouten referred to the action within a record database system, a person ordinary skilled in the art can further apply these rule-based methods to another system such as the system from the teaching combination of Abbondanzio/Polleri based on the following improvements. The combined system can adapt the rules engine and module to identify and apply rules to various condition and actions within a computing system for better selection of an action to be performed.)
Regarding claim 10 depends on claim 9, thus the rejection of claim 9 is incorporated.
Schouten teaches the limitation “The method of claim 9, wherein the one or more rules-based conditions and the given machine-learning based condition are assigned as both needing to be satisfied to result in automatic performance of the one or more computer actions” (paragraph 42 “Generally speaking, a rule can comprise a definition of one or more conditions and an associated one or more actions to be performed upon satisfaction” Schouten discloses a rule comprises a definition of one or more conditions and an associated one or more actions to be performed upon satisfaction of the rule, wherein a person ordinary skilled in the art can further configure the satisfaction of the rule with regard to an action to be performed as well as machine learning model outputs corresponding to an action to be performed that satisfy a metric as disclosed from the teaching by Polleri, as both need to be achieved to initiate the action within the computing system.)
Regarding claim 11 depends on claim 9, thus the rejection of claim 9 is incorporated.
Polleri teaches the limitation “The method of claim 9, wherein the given machine-learning based condition, if satisfied standing alone, results in automatic performance of the one or more computer actions” (paragraph 84 “The monitoring engine can determine if the machine learning model meets the Key Performance Indicators/Quality of Service metrics. Feedback from the monitoring engine can be sent to the model composition engine to provide recommendations to revise the machine learning model”, and paragraph 132 “In certain embodiments, an utterance received as input by digital assistant 406 goes through a series or pipeline of processing steps. ... determining an action to be performed in response to the utterance, causing the action to be performed”. Polleri discloses the monitoring engine can determine if the machine learning model with its output meets the metrics. The machine learning model is deemed to satisfied if they meet the predetermined metric. Polleri further discloses the system receive input from a user, process the input through a series or pipeline of processing steps, which include a machine learning model, and finally output an outcome that need to meet a metric to allow the system to determine and causing an action to be performed in response to the input. A person ordinary skilled in the art can further configured that only one output of the machine learning model is required to meet the corresponding metric to be outputted as the result to determine an action to be performed by the system, wherein the action may correspond to an action within the computing system as disclosed by Abbondanzio based on the teaching combination.)
Claim(s) 17-19 are being rejected under 35 U.S.C. 103 as being unpatentable over Abbondanzio et.al (US 20180167261 A1) in view of Polleri et.al (US 20210081819 A1), further in view of Guo et.al (US 20200210899 A1).
Regarding claim 17,
Polleri teaches the limitation “identifying, for a given machine-learning based condition, one or more criteria for actions that are indicative of the machine-learning based condition” (paragraph 73 “user input can include training metrics. The training metrics help evaluate the performance of the model. Example training metrics can include classification accuracy, logarithmic loss, area under curve, F1 Score, mean absolute error, and mean squared error. The accuracy metric is a ratio of the number of correct predictions divided by the number of predictions made”, and paragraph 96 “At 318, the functionality includes training the machine learning model with predictions judged against QoS/KPIs. In various embodiments, the data can be used to train the machine learning model. The variables of the model can be adjusted based on the output values based on the QoS/KPI metrics”. Polleri discloses a system, wherein the system receives user input comprising various metrics to evaluate the performance of the model. These metrics suggest the one or more criteria within the claim. The output of the machine learning model is evaluated against the metric, wherein the output suggests the machine-learning based condition within the claim.)
Polleri teaches the limitation “determining corresponding instances of data, of a user or organization, based on each of the instances of data being associated with one or more corresponding computer actions that satisfy the one or more criteria” (paragraph 132 “In certain embodiments, an utterance received as input by digital assistant 406 goes through a series or pipeline of processing steps. These steps may include, for example, parsing the utterance, understanding the meaning of the utterance, refining and reforming the utterance to develop a better understandable structure for the utterance, determining an action to be performed in response to the utterance, causing the action to be performed, generating a response to be output to the user responsive to the user utterance”, and paragraph 133 “A digital assistant 406 may use an NLP engine and/or a machine learning model (e.g., an intent classifier) to map end user utterances to specific intents (e.g., specific task/action or category of task/action that the chatbot can perform)”. Polleri discloses a user may provide input to the digital assistant within the system, the digital assistant utilizes machine learning model to provide output corresponding to specific intents such as a task or action that the bot can perform, wherein the machine learning output may be evaluated through a metric as disclosed above.)
Abbondanzio teaches the limitation “receiving one or more instances of user interface input directed to an automation interface, the one or more instances of user interface input defining” (paragraph 22 “Still further embodiments of the method may include receiving, via an interface of the mobile computing device, user-input selecting an action from among the plurality of displayed actions” Abbondanzio discloses a user interface that allow user to provide input such as a selection of an action to be performed.)
Abbondanzio teaches the limitation “one or more computer actions to be performed automatically in response to satisfaction of one or more action conditions” (paragraph 16 “Embodiments of the method display actions that are predetermined to be responsive to the condition. Such an action may be predetermined to be responsive to a condition though a stored association of the action to the condition”, and paragraph 23 “In other embodiments, the method may further comprise automatically invoking a given action from among the plurality of actions.” Abbondanzio discloses the method display and automatically invoke an action to be performed, wherein the action is responsive to a condition. A user ordinary skilled in the art would have been able to incorporate the various features from the teaching by Polleri such as the interface by Abbondanzio may display various actions and after the user select the action, the action may be provided to a bot system by Polleri to perform machine learning process and evaluate the output against a metric. The bot system may indicate if the output satisfies the metric, such that the action corresponding with the output may be automatically invoked to be performed.)
Polleri teaches the limitation “one or more action conditions, including the machine-learning based condition” (paragraph 211 “A bot may use a natural language processing (NLP) engine ... a machine learning based NLP engine may learn to understand and categorize the natural language conversation from the end user and to extract necessary information from the conversation to be able to take precise actions”, and paragraph 272 “A model execution system 1018 may access the trained machine-learning models 1015, provide and format input data to the trained models 1015 ... and determine the predicted outcomes based on the execution of the models. The outputs of the trained models 1015 may be provided to client devices 1050 or other output systems via the API 1012 and/or user interface components”. Polleri discloses the bot uses a machine learning based NLP engine to process information to take precise action, wherein the machine learning engine comprises trained machine-learning models that provide outputs corresponding with the action of the bot.)
Polleri teaches the limitation “based on the one or more instances of user interface input being from the user or an additional user of the organization, and based on the machine-learning based condition being included in the defined one or more action conditions” (paragraph 50 “A model composition engine 132 can be executed on one or more computing systems (e.g., infrastructure 128). The model composition engine 132 can receive inputs from a user 116 through an interface 104. The interface 104 can include various graphical user interfaces with various menus and user selectable elements. The interface 104 can include a chatbot (e.g., a text based or voice based interface). The user 116 can interact with the interface 104 to identify one or more of: a location of data, a desired prediction of machine learning application, and various performance metrics for the machine learning model.” Polleri discloses a user interface that a user may interact with to provide input and the bot system may perform a task or action based on utilizing one or more machine learning model outputs and metrics.)
Polleri teaches the limitation “using the tailored machine-learning model in determining whether the one or more action conditions are satisfied in determining whether to automatically perform the one or more computer actions” (paragraph 54 “The monitoring engine 156 can receive the results of the model execution engine 108 and compare the results with the performance characteristics (e.g., KPI/QoS metrics 160). The monitoring engine 156 can use ground truth data to test the machine learning application 112 to ensure the model can perform as intended”, and paragraph 176 “After the end user intent is determined based on the content by message processor 550, the determined intent (and the parameters associated with the intent) may be sent to an action engine 560. Action engine 560 may be used to determine an action to perform based on the intent”. Polleri discloses The monitoring engine can receive the results such as output of a machine learning model and evaluate the output against performance characteristics such as various metrics to determine if the output is satisfy, wherein the bot system relies on the output of the machine learning model evaluated against the metric to determine the best action to perform based on user input, wherein the action may be a computing action as the teaching by Polleri is combine with the teaching by Abbondanzio.)
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 DUY TU DIEP whose telephone number is (703)756-1738. The examiner can normally be reached M-F 8-4:30.
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, Alexey Shmatov can be reached at (571) 270-3428. 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.
/DUY T DIEP/Examiner, Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123