DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
The office action is being examined in response to the application filed by the applicant on 20 August 2025.
Claims 1-20 are pending and have been examined.
This response amends claims 1, 10, and 17.
This action is made FINAL.
Response to Arguments
35 U.S.C. § 101 Arguments – Non-Statutory Subject Matter
Applicant’s Remarks, see page 8, filed 8/20/2025, with respect to the rejection of claims 1-9 under 35 U.S.C. § 101 for being directed to non-statutory subject matter have been fully considered and are persuasive. Therefore, the rejection has been withdrawn.
35 U.S.C. § 101 Arguments – Abstract Idea
Applicant’s Remarks, see pages 8-14, filed 8/20/2025, with respect to the rejection of claims 1-20 under 35 U.S.C. § 101 have been fully considered and are not persuasive.
On pages 8, the Applicant argues with respect to 35 U.S.C. § 101, Step 2A, Prong 1, and asserts that that the rejection is improper and should be withdrawn due to the inclusion of structural elements. The Examiner respectfully disagrees. This argument is not germane to 35 U.S.C. § 101, where the mere inclusion of the structures in the wording of the rejection is not a defect according to MPEP 2106.07. The claim still recites instructions to implement the function of the abstract idea via the structure elements, e.g. the structure elements are merely the tools used to implement the instructions, adding the words “apply it.”
On pages 9-13 of the arguments, the Applicant argues with respect to 35 U.S.C. § 101, Step 2A, Prong 2. The Applicant asserts that a real-time data store and the write-side and read-side portions of the feature generation system provide advances to the speed of data storage and retrieval from databases, the efficiency and speed of generating activity features for a machine learning model, and that the write-side portion is not recited at a high level of generality as it is specifically designed to collect a data stream and join it with mapped attribute data, thereby setting forth an improvement to the technology or technical field, which do not merely confine the use of the abstract idea to a technological field.
The Examiner respectfully disagrees. "Claiming the improved speed or efficiency inherent with applying the abstract idea,” while merely utilizing the structure elements as tools, “does not integrate a judicial exception into a practical application or provide an inventive concept,” (MPEP 2106.05(f) and MPEP 2016.05(f)(2)). Thus, the structural elements are recited with a high level of generality, such that the claims fail to differentiate the functions from being performed by any reasonable implementation of any write-side or read-side portion of any feature generation system utilizing any real-time data store, without identifying succinct metes and bounds that might identify the structures otherwise. Since the additional elements are merely applied to increase the speed and efficiency, they do not improve the function of a computer, or any technology or technical field. The claims generally link the use of the abstract ideas to the particular technological environment,
The Applicant’s argues on page 14 regarding 35 U.S.C. § 101 Step 2B, presenting the previous assertions from steps 2A, Prongs 1 and 2, and the Examiner respectfully disagrees for the same reasons above. Therefore, the claims do not integrate the judicial exception into a practical application or amount to significantly more than the abstract idea (MPEP 2106.05(a), (e), (f), and (h)).
On pages 13 and 14, the Applicants asserts that the claims are patent eligible and respectfully requests that the rejection should be withdrawn. The Examiner respectfully disagrees and the 35 U.S.C. § 101 rejection is maintained. Please find an updated rejection for 35 U.S.C. § 101 below, reflecting the amendments.
35 U.S.C. § 103 Arguments
Applicant’s arguments, see pages 14-15, filed 8/20/2025, with respect to 35 U.S.C. § 103 in claims 1-4, 6, 9-10, and 14-20, have been fully considered and are persuasive. Therefore, the rejections for claims 1, 11, and 17 have been withdrawn. However, upon further consideration, a new grounds of rejection is made in view of a newly found prior art reference necessitated by the amended claims.
On page 15, the Applicant asserts that neither of the prior art disclosures of Liu or Bhardwaj, taken alone or in combination, mention joining data from two different sources, nor do they perform the function in the manner claimed, i.e. utilize a write-side portion of a feature generation system to record events into one data store and join those events with attributes from a second data store, where the attributes map entities to events of the stream. Otherwise, the Applicant’s arguments fail to provide any other arguments as to how the prior art fails to distinguish the claims from the references.
Since the write-side and read-side portions of the feature generation system are newly amended claim limitations that were not previously presented, the new 35 U.S.C. § 103 rejections for the claims are presented below.
On page 15, the Applicant argues with respect to 35 U.S.C. § 103 for the dependent claims. The Applicant asserts that the dependent claims are patentable because they depend directly or indirectly on the independent claims, which the Applicant asserted were Allowable. The Examiner respectfully disagrees. The Applicant is merely relying on the independent claims and fails to argue much more than a general allegation of patentability without pointing out how the language of the claims patentably distinguish them from the references. Sans the amended claim language in the Independent and dependent claims, the Applicant’s arguments for the dependent claims fail to comply with 37 CFR 1.111(b) for this same reason.
Accordingly, based on arguments the and the detailed analysis above, the 35 U.S.C. § 103 rejection is withdrawn. Please find the updated 35 U.S.C. § 103 rejection below to reflect new art for the amended claim language in the independent claims necessitated by claim amendments.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the systems and methods recited by the claims are directed to abstract ideas without significantly more.
Independent Claims
Regarding Claim 1: Step 2A Prong 1: The claim recites the following function: record events, obtain attributes, join attributes and events, produce event data, receive a request for a feature and a timestamp, determine a data access mechanism, retrieve event data, compute features, input data, and provide computed feature, which are abstract ideas in the category of “certain methods of organizing human activity,” because the claim utilizes the functions above to manipulate user/event/entity/attribute activity data to compute and provide a feature according to the user’s interface activities and drive personal behaviors or relationships, or interactions between people (MPEP 2106.05(a)(2)(II)). These functions are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind,” i.e. historically human tasks.
Step 2A Prong 2: The claim recites the following additional computing elements: A system, processor, memory, a first data store, real-time data store, and a second data store. These limitations are recited in the claims and disclosed in the specification at a high level of generality, i.e. they are disclosed as general-purpose computing structures. This amounts to “apply it,” such that the claims are mere instructions to implement abstract ideas on general purpose computing structures.
The claim recites send, receive, retrieve, provide, and input data. The specification does not reveal that the core of the invention is directed to advances in sending, receiving, retrieving, inputting, or providing data. Instead, the specification is focused on the nature and the timing (real-time vs batch processing) of the data being manipulated, i.e., the descriptive nature of the data (MPEP 2106.05(e)).
The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
Insofar as the claim recites a computing device that executes an application, the specification discloses the computing device is simply a generic computing structure at a high level of generality. Instructions to apply an abstract idea on a general-purpose structure is not a practical application (MPEP 2106.05(f)).
The claims recite a machine learning model, data access mechanism (database query), feature configuration (computer code), a write-side portion and read-side portion of a feature generation system, a feature computation algorithm, and an interface (software), i.e. software and action-based computer code, recited at a high level of generality. The claims recite the functions in terms of intended uses and intended results, i.e. what the function does and what the function returns, without describing how the system performs the functions, i.e. the claim puts no limitations as to how these steps are performed. A claim that merely recites record a stream of data, obtain mapped attributes, receive a request for data, determine data, retrieve data, compute a feature using the data, and provide the feature to answer the request, does not describe how the system performs the generally recited functions. These receiving, requesting, determining, computing, providing, invoking, sending, and receiving functions are disclosed as instructions executed on a generic computing structure that utilize the general-purpose machine learning models, queries, configurations, or algorithms as tools to implement the abstract ideas, without limitations as to how these functions are performed i.e. adding the words “apply it.” Further, the data generation functions performed by the read and write-side portions are also performed by general purpose structures which does not reveal advances to the functioning of a computer, or to any other technology or technical field (MPEP 2106.05(a)). In fact, the claim only recites the steps to achieve the outcomes. Instructions to generally link a judicial exception to a particular field use is not a practical application (MPEP 2106.05(h)). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claim 10 Step 2A Prong 1: Claim 10 recites: record a user events, obtain mapped attributes, join attributes and events, receive user interface activity, send output request, send request for feature and timestamp, read configuration, determine a data access mechanism, a time window, and an algorithm (matched from the configuration), receive processed event data, compute feature, input event and attribute data, respond to request, provide feature, respond to output request, generate output, input feature, provide output, generate user interface output, respond to user activity, and send interface output, which are abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim manipulates user activity features, a task historically provided by humans, i.e. automating advanced statistical analysis, according to the user’s interface activities to drive human interactions(MPEP 2106.05(a)(2)(II)). These functions are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind.” The claim puts no limitations as to how these steps are performed (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim recites data structures, i.e. general-purpose data storage or database structure which are not an abstract idea.
The claim also recites the following send and receive functions: record (receive) a user events, obtain mapped attributes, receive user interface activity, send output request, send request for feature and timestamp, receive processed event data, input event and attribute data, respond to request, provide feature, respond to output request, input feature, provide output, generate user interface output, respond to user activity, and send interface output. The specification does not reveal that the core of the invention is directed to advances in sending, receiving, retrieving, or providing data, to advances in accessing databases, the way or speed that data is stored or retrieved from databases, database structures, or advances in or the invention of a database architecture. These limitations are disclosed at a high level of generality. Instead, the specification is focused on the nature and the timing (real-time vs batch processing) of the data being received, requested, retrieved, provided, matched, or computed for the recommender system – i.e., the descriptive nature of the data (MPEP 2106.05(e)). These limitations are not abstract ideas and do not amount to a practical application of an abstract idea.
The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The claim recites the feature configuration, which is one feature configuration of a library of feature configurations that are used to determine the particular algorithms to be executed to generate the features requested by the machine learning model along with the set of inputs needed by the computation algorithms to generate those features. Therefore, this limitation is a both a predefined data container and a characterization of the data container, not an abstract idea, nor limitation that carries patentable weight in the claim. The limitation cannot be relied on to integrate the abstract idea into a practical application because they (the individual and the library) are non-functional descriptive materials – they do not positively recite any additional functions that limit the claims or the structures of the claims, they are limited by the user activities.
The client device is disclosed as contained in a computing device, which is disclosed a generic computer structure described at a high level of generality. Instructions to apply an abstract idea on a general-purpose structure is not a practical application (MPEP 2106.05(f)).
The claim also recites an application system (API), machine learning model, user interface (software), feature generation system (algorithm), write and read side portions of the feature generation system, data access mechanism (executable queries, i.e., algorithm), and feature computation algorithm. i.e. software and action-based computer code, recited at a high level of generality. The claim recites the functions in terms of intended uses and intended results, i.e. what the function does and what the function returns, without describing how the system performs the functions, i.e. the claim puts no limitations as to how these steps are performed. A claim that merely recites record (receive) a user events, obtain mapped attributes, receive user interface activity, send output request, send request for feature and timestamp, receive processed event data, input event and attribute data, respond to request, provide feature, respond to output request, input feature, provide output, generate user interface output, respond to user activity, and send interface output, does not describe how the system performs the generally recited functions. These functions are disclosed as instructions executed, that utilize the general-purpose machine learning models, queries, configurations, or algorithms as tools to implement the abstract ideas, without limitations as to how these functions are performed i.e. adding the words “apply it.” Further, the data generation functions performed by the read and write-side portions are also performed by general purpose structures which does not reveal advances to the functioning of a computer, or to any other technology or technical field (MPEP 2106.05(a)). In fact, the claim only recites the steps to achieve the outcomes. Instructions to generally link a judicial exception to a particular field use is not a practical application (MPEP 2106.05(h)). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claim 17 Step 2A Prong 1: Claim 17 recites: record event, obtain mapped attributes, join attributes and events, produce processed event data, receive output request, send a data and timestamp data, read configuration for feature, determine a data access mechanism, determine a time window, determine an algorithm, i.e. choose a matching algorithm based on the configuration, retrieve processed event data within window, retrieve attributed data for event data, compute the feature, provide feature, respond to request for output, generate an output, and provide the output, all dependent on the particular user interface activities, which are abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim manipulates user activity features according to the user’s interface activities to drive human interactions(MPEP 2106.05(a)(2)(II)). These functions are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind,” (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim recites the real-time data store, which is a general-purpose data storage or database structure and is not an abstract idea.
The claim also recites the following send and receive functions: record (receive) event, obtain attributes, receive output request, send data and timestamp data, retrieve processed event data within window, retrieve attributed data for event data, provide feature, respond to request for output, and provide the output. The specification does not disclose that the core of the invention is directed to advances in sending, receiving, retrieving or providing data, to advances in accessing databases, the way or speed that data is generated, stored, or retrieved in association with databases (since the system utilizes commercially available systems as disclosed in the claim 1 analysis for the read and write-side portions), database structures, or advances in or the invention of a database architecture. These limitations are disclosed at a high level of generality without limitations as to how these steps are performed. Instead, the specification is focused on the nature (i.e. manipulating data like attributes mapped to entities in the event stream and the user interface events in a stream), and the timing (real-time vs batch processing) of the data being received, requested, retrieved, provided, matched, or computed for the recommender system – i.e., the descriptive nature of the data (MPEP 2106.05(e)). These limitations are not abstract ideas and do not amount to a practical application of an abstract idea.
The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The claim recites the feature configuration, which is one feature configuration of a library of feature configurations that are used to determine the particular algorithms to be executed to generate the features requested by the machine learning model along with the set of inputs needed by the computation algorithms to generate those features. Therefore, this limitation is a both a predefined data container and a characterization of the data container, not an abstract idea, nor limitation that carries patentable weight in the claim. The limitation cannot be relied on to integrate the abstract idea into a practical application because they (the individual and the library) are non-functional descriptive materials – they do not positively recite any additional functions that limit the claims or the structures of the claims, they are limited by the user activities.
The client device and data stores are general purpose computing structures disclosed at a high level of generality. Instructions to apply an abstract idea on a general-purpose computing structure is not a practical application (MPEP 2106.05(f)).
The claim also recites an application system (API), machine learning model, user interface (software), feature generation system, data access mechanism (executable queries, i.e., algorithm), read and write-side of a feature generation system, and feature computation algorithm, where these limitations are recited at a high level of generality, and are not abstract ideas. These software functions are disclosed as generic software, API’s, machine learning models, user interface software, executable queries, or algorithms. The claim recites the functions in terms of intended uses and intended results, i.e. what the function does and what the function returns, without describing how the system performs the functions, i.e. the claim puts no limitations as to how these steps are performed. A claim that merely recites record event, obtain mapped attributes, join attributes and events, produce processed event data, receive output request, send a data and timestamp data, read configuration for feature, determine a data access mechanism, determine a time window, determine an algorithm, i.e. choose a matching algorithm based on the configuration, retrieve processed event data within window, retrieve attributed data for event data, compute the feature, provide feature, respond to request for output, generate an output, and provide the output, does not describe how the system performs the generally recited functions. These functions are disclosed as instructions executed on a generic computing structure that utilize the general-purpose machine learning models, queries, configurations, or algorithms as tools to implement the abstract ideas, without limitations as to how these functions are performed i.e. adding the words “apply it.” Further, the data generation functions performed by the read and write-side portions are also performed by general purpose structures which does not reveal advances to the functioning of a computer, or to any other technology or technical field (MPEP 2106.05(a)).
These types of limitations merely confine the use of the abstract idea to a particular technological environment and thus fail to add an inventive concept to the claim (MPEP 2106.05(h)). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Dependent Claims
Regarding Claim 2: Step 2A Prong 1: Claim 2 recites: compute a user activity feature, perform a data aggregation, filtering, and a grouping function, all abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim manipulates user activity features, tasks historically performed by humans using advance statistical analysis, according to the user’s interface activities, to drive human interactions(MPEP 2106.05(a)(2)(II)) . All of which are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind,” since these functions are simply matching data types, or providing data that could be performed by hand or that can be done in the mind. The claim puts no limitations as to how these steps are performed, and the application system is merely a tool used to perform the otherwise mental processes (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The computing device is disclosed as a generic computing structure disclosed with a high level of generality. Instructions to apply an abstract idea on a general-purpose computing structure is not a practical application (MPEP 2106.05(f)).
The specification does not reveal that the core of the invention is directed to advances in data sorting, aggregation, filtering, grouping, or matching techniques, data storage techniques, data providing techniques, proximity or proximity settings or related techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. Instructions to generally link a judicial exception to a particular field use is not a practical application (MPEP 2106.05(h)). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claims 3 Step 2A Prong 1: claim 3 recites: compute a sum, a count, an average, a date comparison, and a probability distribution of a user activity feature, all abstract ideas in the category of “mathematical concepts”, more specifically, “mathematical relationships” and “mathematical calculations” because the claim mathematically transmute the data into the display output (MPEP 2106.05(a)(2)(I)). The claim recitations above and the following recitations: perform an aggregation, and compute an average pooling and a histogram, are abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim utilizes the functions above to perform tasks historically performed by humans, to manipulate user activity features according to the user’s interface activities and drive personal behaviors or relationships, or interactions between people (MPEP 2106.05(a)(2)(II)). These functions are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind.” The claim puts no limitations as to how these steps are performed. The system is merely a tool used to perform the otherwise mental processes (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The system is disclosed as a general-purpose computing structure with a high level of generality. Instructions to apply an abstract idea on a general-purpose structure is not a practical application (MPEP 2106.05(f)).
The specification does not reveal that the core of the invention is directed to advances in data sorting, aggregating, computing, summing, counting, averaging, comparing, average pooling, or matching techniques, histogram or probability distribution techniques, data storage techniques, data providing techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. Instructions to generally link a judicial exception to a particular field use is not a practical application (MPEP 2106.05(h)). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claim 4 Step 2A Prong 1: Claim 4 recites: obtain instances of event data and query to find matches to the user activities, which are abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim manipulates user activity features according to the user’s interface activities to drive human interactions (MPEP 2106.05(a)(2)(II)). These are also an abstract idea in the category of “mental processes” or “things that can be performed in the human mind.” The claim puts no limitations as to how these steps are performed, where the computing device and network-based service are merely tools used to perform the otherwise mental processes (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The system is disclosed as a general-purpose computing structure with a high level of generality. Instructions to apply an abstract idea on a general-purpose structure is not a practical application (MPEP 2106.05(f)).
The specification does not reveal that the core of the invention is directed to advances in data sorting, aggregating, querying, sending, receiving, obtaining, or matching techniques, data storage techniques, data providing techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcome described at a high level of generality without placing any limitations to how the application functions. This type of limitation merely confines the use of the abstract idea to a particular technological environment without a practical application (an application to obtain data via a query) and thus fails to add an inventive concept to the claims. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. See MPEP 2106.05(h). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claim 5 Step 2A Prong 1: Claim 5 recites: the real-time data store, which is a general data structure, is not an abstract idea and cannot be relied upon to integrate the judicial exception into a practical application. Additionally, claim 5 recites the data store is arranged according to a schema, which is a data store characterization, thus is non-functional descriptive information, is not an abstract idea, and carries no patentable weight. Lastly, claim 5 recites a feature type associated with the request that defines the data store from a library of data stores that are utilized for each feature type, which is also a data store characterization linked to the input, is not an abstract idea, carries no patentable weight and cannot be relied upon to integrate the judicial exception into a practical application. There are no further limitations, functions, or elements that can be relied on to integrate the judicial exceptions into practical applications.
Regarding Claim 6 Step 2A Prong 1: Claim 6 recites: obtain the attribute data and perform a sequential lookup according to the entity identifier, which are abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim manipulates user activity features according to the user’s interface activities to drive human interactions (MPEP 2106.05(a)(2)(II)). All of which are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind,” since these functions are simply matching data types, or providing data that could be performed by hand or that can be done in the mind. The claim puts no limitations as to how these steps are performed, and the application system is merely a tool used to perform the otherwise mental processes (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The system is disclosed as a general-purpose computing structure with a high level of generality. Instructions to apply an abstract idea on a general-purpose structure is not a practical application (MPEP 2106.05(f)).
The specification does not reveal that the core of the invention is directed to advances in data lookup, obtaining data or data matching techniques, data storage techniques, data providing techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcome described at a high level of generality without placing any limitations to how the application functions. This type of limitation merely confines the use of the abstract idea to a particular technological environment without a practical application (an application to obtain data via a query) and thus fails to add an inventive concept to the claims. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. See MPEP 2106.05(h). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions
Regarding Claim 7: Step 2A Prong 1: Claim 7 recites: define a maximum value of a time window as N days prior to and including a day of the request timestamp, define N to be a positive integer, which are abstract ideas in the category of “mathematical concepts”, more specifically “mathematical relationships,” “mathematical formulas or equations,” and “mathematical calculations” or relationships (MPEP 2106.05(a)(2)(I)). These are also abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim manipulates user activity features according to the user’s interface activity to drive human interactions (MPEP 2106.05(a)(2)(II)). These are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind,” because these functions could be performed by hand using mathematical calculations can be done in the mind, and the claim puts no limitations as to how these steps are performed, where the computing device and network-based service are merely tools used to perform the otherwise mental processes (MPEP 2016.04(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The system is disclosed as a general-purpose computing structure with a high level of generality. Instructions to apply an abstract idea on a general-purpose structure is not a practical application (MPEP 2106.05(f)).
The specification does not reveal that the core of the invention is directed to advances in data defining, mathematical techniques, data storage techniques, data providing techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. Instructions to generally link a judicial exception to a particular field use is not a practical application (MPEP 2106.05(h)). There are no further limitations, functions, or elements that can be relied on to integrate the judicial exceptions into practical applications. The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claims 8 Step 2A Prong 1: Claim 8 recites: compute a difference between the timestamps to be less than 100 milliseconds, which is an abstract idea in the category of “mathematical concepts”, more specifically, “mathematical relationships,” “mathematical formulas and equations,” and “mathematical calculations” or relationships because the claim mathematically puts a limit on the calculated timestamps between the request and the user activity feature (MPEP 2106.05(a)(2)(I)). These are also an abstract idea in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim manipulates user activity features according to the user’s interface activities to drive human interactions (MPEP 2106.05(a)(2)(II)). Lastly, this is also an abstract idea in the category of “mental processes” or “things that can be performed in the human mind,” where assigning data value limits when calculating the timestamp differences data can be done in the mind (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The system is disclosed as a general-purpose computing structure with a high level of generality. Instructions to apply an abstract idea on a general-purpose structure is not a practical application (MPEP 2106.05(f)).
The specification does not reveal that the core of the invention is directed to advances in data defining, mathematical techniques, data storage techniques, data providing techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. Instructions to generally link a judicial exception to a particular field use is not a practical application (MPEP 2106.05(h)). The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claims 9 Step 2A Prong 1: Claim 9 recites: the real-time data store, which is a general data structure, is not an abstract idea and cannot be relied upon to integrate the judicial exception into a practical application. Additionally, claim 9 recites the location of the real-time data store’s, which is merely a URL or path that a query travels to in order to get data, a data pointer comprised in the data access mechanism, within the query, therefore, it is a characterization of the query, and is not an abstract idea. Lastly, claim 9 recites a that the query is formatted so that it can be executed against the real-time data store. This formatting is also a characterization of the query based on the format of the real-time data store, and is not an abstract idea. Query characterizations carry no patentable weight and cannot be relied upon to integrate the judicial exception into a practical application. There are no further limitations, functions, or elements that can be relied on to integrate the judicial exceptions into practical applications.
Regarding Claims 11-16: Step 2A Prong 1: These claims each recite: use model output, driven by particular event data input, to configure various recommendations for output to the user’s interface, which are abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior, relationships, or interactions between people” because the claim utilizes the function to manipulate user activity features according to the user’s interface activities to drive human interactions(MPEP 2106.05(a)(2)(II)). Thes are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind,” since these functions are simply matching data types, or providing data that could be performed by hand or that can be done in the mind. The claim puts no limitations as to how these steps are performed, and the application system is merely a tool used to perform the otherwise mental processes (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The specification does not reveal that the core of the invention is directed to advances in data gathering, API’s, machine learning model architecture, configuring algorithms, algorithm architecture, data storage techniques, data providing techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. Instructions to generally link a judicial exception to a particular field of use is not a practical application (MPEP 2106.05(h)). There are no other claim limitations that can be relied on to integrate the judicial exceptions into a practical application. The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
Regarding Claims 18-20: Step 2A Prong 1: The claims recite: compute the requested user activity feature for particular event data that matches a particular identifier or attribute associated with particular interface activities, which are abstract ideas in the category of “certain methods of organizing human activity,” more specifically “managing personal behavior or relationships or interactions between people” because the claim to manipulates user activity features according to the user’s activities to drive human interactions(MPEP 2106.05(a)(2)(II)). All of which are also abstract ideas in the category of “mental processes” or “things that can be performed in the human mind.” The claim puts no limitations as to how these steps are performed, and the application system is merely a tool used to perform the otherwise mental processes (MPEP 2106.05(a)(2)(III)).
Step 2A Prong 2: The claim also recites data characterizations, which are non-functional descriptive information, and carry no patentable weight.
The specification does not reveal that the core of the invention is directed to advances in data gathering, API’s, machine learning model architecture, configuring algorithms, algorithm architecture, data storage techniques, data providing techniques, data structures, or improvements to the characterization of data. In fact, the claims only recite the steps to achieve the outcomes, without describing how the system performs the functions. Instructions to generally link a judicial exception to a particular field of use is not a practical application (MPEP 2106.05(h)). There are no other claim limitations that can be relied on to integrate the judicial exceptions into a practical application. The claim as a whole, while looking at additional elements individually and in combination, do not integrate the judicial exceptions into a practical application.
Step 2B: The analysis above is commensurate with the analysis for Step 2B, such that the same additional elements taken individually and in combination do not result in the claim, as a whole, amounting to significantly more than the judicial exceptions.
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 1-4, 6, 9-10, and 14-20 are rejected under 35 U.S.C. 103 as being unpatentable over Liu, US20150006295A1, in view of Bhardwaj, US20210035046A1, and in further view of LaBorde, US10043591B1.
Regarding Claim 1: Liu discloses: A system comprising: a processor; and a memory coupled to the processor, wherein the memory comprises instructions that when executed by the processor cause the processor to: [0102] (a computer system with a processor), [0103] (a memory), [0116] (instructions comprised in memory are executed to cause the processor to perform);
by a feature generation system, record a stream of user interface events into a first data store, wherein the first data store comprises a real-time data store; [0050-0051] (data pipeline to collect raw tracking data from user behaviors and interactions, i.e. record a stream of user interface events), [0105-0106] (obtain user interaction data), [0105] (stores events in memory 924, i.e. a first data store), [0020] “the recommendations are computed and applied in real-time,” (such that recommendations computed and applied in real-time must imply real-time data storage);
obtain, attributes that map to entities identified in particular events of the stream; [0109] (obtain attributes), [0108] (attributes map to user interface events);
join the attributes with corresponding events in the stream to produce processed event data; [0051] (raw tracking data), [0020] (tracking data is joined with attributed data), [0053] (information is used to provide features);
a machine learning model; [0107 and 0111] (Machine learning model)
receive a user activity feature, [0107 and 0111] (Machine learning model receives user activity features associated with user activities and user attributes);
using a feature configuration associated with the requested user activity feature determine a data access mechanism, a time window determined based on the request timestamp, and a feature computation algorithm; [0062] (data is accessed according to user’s activities associated with interest segments and categorical content, i.e. the configuration of the features), [0059] ”time window,” and [0108] (different machine learning models containing feature appropriate algorithmic calculations for different feature configurations);
using the data access mechanism, retrieve, from the processed event data, a plurality of instances of event data that each comprise: a user identifier, an event identifier associated with the user identifier, an entity identifier associated with the event identifier, and attribute data associated with the instance of event data; [0050] (data pipeline to collect dynamic actions), [0053] (plurality of instances of event data), [0059] (data collected according to “time windows”), [0020] “the recommendations are computed and applied in real-time,” [0079] (retrieve features, i.e. processed event data);
compute the requested user activity feature using the retrieved plurality of instances of event data and the retrieved attribute data as inputs to the feature computation algorithm; and [0053]
responsive to the request, provide the computed user activity feature to the machine learning model; [0053-0054]
Where Liu does not disclose, Bhardwaj teaches:
from a second data store; [0054] (distributed cloud computing system with distributed memory systems);
receive, from a machine learning model, a request for a user activity feature and a request timestamp; [0021-0023] (a request for a feature and timestamp are triggered by a user executing a user interface activity, where the machine learning model is triggered to send a request for the feature, the request received by the system; this occurs in real-time), [0015] (event parameters include a timestamp);
an event timestamp within the time window; [0015]
a time window defined by the request timestamp; [0047] “event data may be grouped based on time windows (e.g., based on the timestamp of each event).”
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query data collected in databases and present social media and employment recommendations to user’s, utilizing collaborative filtering and machine learning models, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Where Liu does not disclose and Bhardwaj does not teach, LaBorde teaches:
by a write-side portion of a feature generation system, recording a stream of events [Column 10, Lines 59-68] (Command Query Responsibility Segregation, CQRS, where the command and query are segregated such that the stream of events is stored and may be reconstructed by the command side, and the query side is used to query for data to input into machine learning models; streams of user events are recorded in the command side, i.e. the command side is the write-side portion in the feature generation system);
by a read-side portion of the feature generation system; [Column 10, Lines 59-68] (Command Query Responsibility Segregation, CQRS, where the command and query sides are segregated such that the stream of events is received, stored, and may be reconstructed by the command side, i.e. the write-side, and the query side, i.e. the read-side, is used to query for data to input into machine learning models and return results), [Column 11, Lines 50-54] “providing data input into the system and the predictive analytic outputs of the proprietary models that consume this data. Data generated via these work flows executed in the client application in the regular course of its use;”
A first and second data store [Column 10, Lines 16-20, and Figure 4] (at least two different databases utilized for storage and retrieval of data in the CQRS, Command and Query portions of the system).
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query data collected in databases and present social media and employment recommendations to user’s, utilizing collaborative filtering and machine learning models, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Regarding Claim 2: Liu discloses: The system of claim 1, wherein the computing device computes the user activity feature by performing one or more of an aggregation, a filtering, or a grouping, of the attribute data. [0053] (aggregation), [0091] (filtering), [fig. 3, 314] (grouping).
Regarding Claim 3: Liu discloses: The system of claim 2, wherein performing the aggregation comprises computing, over the time window one of a sum, a count, an average, a date comparison, an average pooling, a histogram, or a probability distribution, on the attribute data. [0058] “impressions and clicks may be aggregated daily or with some other period over some length of time. As a monthly average, for example, there may have been 7 million impressions served to 1.3 million viewers, with 4,500 clicks received.”
Regarding Claim 4: Liu discloses: The system of claim 1, wherein the instructions further cause the processor to obtain the plurality of instances of event data by querying the real-time data store [0050] (data pipeline to collect dynamic actions), [0053] (plurality of instances of event data), [0059] (data collected according to “time windows”), [0020] “the recommendations are computed and applied in real-time.”
Regarding Claim 6: Liu discloses: The system of claim 1,
Where Liu does not disclose, Bhardwaj teaches: wherein the instructions further cause the processor to obtain the attribute data by performing a sequential lookup on a key-value store using the entity identifier as a key. [0018] “Each of the click-stream events may also include a timestamp as well as one or more parameters (referred to herein as "event_params") associated with the event. The event_params may be expressed as key-value pairs of attributes related to the event_type… each event type has an associated timestamp and one or more listing category” and [0023] “Upon receipt of the request and the parameters, the recommendation engine associated with the host platform 120 performs one or more queries of the listings database.”
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and machine learning for social media and employment networks with database querying to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system utilizing key-value pair querying.
Regarding Claim 9: Liu discloses: The system of claim 1,
Where Liu does not disclose Bhardwaj teaches: wherein the data access mechanism, the real-time data store comprises a query in a format that can be executed against the real-time data store and a location of the real-time data store. [0041]
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query databases and present social media and employment recommendations to user’s, utilizing collaborative filtering and machine learning models, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Regarding Claim 10: Liu discloses: A method comprising:
by a feature generation system, record a stream of user interface events into a first data store, wherein the first data store comprises a real-time data store; [0050-0051] (data pipeline to collect raw tracking data from user behaviors and interactions, i.e. record a stream of user interface events), [0105-0106] (obtain user interaction data), [0105] (stores events in memory 924), [0020] “the recommendations are computed and applied in real-time,” (recommendations computed and applied in real-time imply real-time data storage);
obtain, attributes that map to entities identified in particular events of the stream; [0109] (obtain attributes), [0108] (attributes map to user behavior events);
join the attributes with corresponding events in the stream to produce processed event data; [0051] (raw tracking data), [0020] (tracking data is joined with attributed data), [0053] (information is used to provide behavioral and profile features);
receiving, at an application system that uses output of a machine learning model to configure a user interface in response to user activity from a client device, data that indicates a user interface activity in the application system; [0107-0114] (API receives user activity user activities, where the machine learning model configures a user interface output in response to client device user activities);
sending, a request for a user activity feature, to a feature generation system; [0072-0073]
reading, by the feature generation system, a feature configuration associated with the requested user activity feature; [0088]
determining, by the feature generation system, based on the feature configuration, a data access mechanism for the real-time data store, a time window defined by the request timestamp, and a feature computation algorithm; [0062] (data is accessed according to user’s activities associated with interest segments and categorical content, i.e. the configuration of the features), [0059] (data collected according to ”time windows”), and [0108] (different machine learning models containing feature appropriate algorithmic calculations for different feature configurations);
retrieving, from the real-time data store by the feature generation system using the data access mechanism, a plurality of instances of the processed event data that each comprise: a user identifier associated with the user interface activity, an event identifier associated with the user identifier, an entity identifier associated with the event identifier, an event timestamp within the time window, and attribute data associated with the plurality of instances of event data; [0050] (data pipeline to collect dynamic actions), [0053] (plurality of instances of event data), [0059] (data collected according to “time windows”), [0020] “the recommendations are computed and applied in real-time,” (Examiner notes that real-time computations requires real-time data), [0079] (retrieve features, i.e. processed event data);
computing, by the feature generation system, the requested user activity feature using the plurality of instances of event data and the attribute data as inputs to the feature computation algorithm; [0053]
responsive to the feature request, by the feature generation system, providing the computed user activity feature to the machine learning model; [0053-0054]
responsive to the request for model output, by the machine learning model, generating a model output using the computed user activity feature as an input, and providing the model output to the application system; [0053-0054].
generating, by the application system, user interface output based on the model output; and [0090];
responsive to the user interface activity, by the application system, sending the user interface output to the client device. [0082].
Where Liu does not disclose, Bhardwaj teaches:
from a second data store; [0054] (distributed cloud computing system with distributed memory systems);
sending, by the application system, to the machine learning model, a request for model output; [0021] (a request for a recommendation, i.e. a request for output), [0023] (an API sends a query for model output to the recommendation engine);
sending, by a machine learning model, a request for a user activity feature and a request timestamp; [0021-0023] (a request for a feature and timestamp are triggered by a user executing a user interface activity, where the machine learning model is triggered to send a request for the feature and timestamp, the request received by the system in real-time), [0015] (event parameters include a timestamp);
and a request timestamp; [0015] “timestamp;”
a time window defined by the request timestamp, [0047] “event data may be grouped based on time windows (e.g., based on the timestamp of each event);”
an event timestamp within the time window; [0015]
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query databases and present social media and employment recommendations to user’s, utilizing collaborative filtering, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Where Liu does not disclose and Bhardwaj does not teach, LaBorde teaches:
by a write-side portion of a feature generation system, recording a stream of events [Column 10, Lines 59-68] (Command Query Responsibility Segregation, CQRS, where the command and query are segregated such that the stream of events is stored and may be reconstructed by the command side, and the query side is used to query for data to input into machine learning models; streams of user events are recorded in the command side, i.e. the command side is the write-side portion in the feature generation system);
by a read-side portion of the feature generation system; [Column 10, Lines 59-68] (Command Query Responsibility Segregation, CQRS, where the command and query sides are segregated such that the stream of events is received, stored, and may be reconstructed by the command side, i.e. the write-side, and the query side, i.e. the read-side, is used to query for data to input into machine learning models and return results), [Column 11, Lines 50-54] “providing data input into the system and the predictive analytic outputs of the proprietary models that consume this data. Data generated via these work flows executed in the client application in the regular course of its use;”
A first and second data store [Column 10, Lines 16-20, and Figure 4] (at least two different databases utilized for storage and retrieval of data in the CQRS, Command and Query portions of the system).
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query data collected in databases and present social media and employment recommendations to user’s, utilizing collaborative filtering and machine learning models, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Regarding Claim 14: Liu discloses: The method of claim 10, wherein the plurality of instances of event data each comprise a connection invitation, and the application system uses the model output to filter a connection invitation portion of the user interface output. [0051] (event data comprises connection invitations) and [0089] (recommendation to connect requires collaborative filtering).
Regarding Claim 15: Liu discloses: The method of claim 10, wherein an instance of the plurality of instances of event data comprises one of a user interaction with a message, a profile view, a page view, or a search query and the application system uses the model output to configure a search suggestion portion of the user interface output [0063] “user interactions with advertising, recommendations, content and each other,” and [0093] (filter a job search recommendation).
Regarding Claim 16: Liu discloses: The method of claim 10, wherein an instance of the plurality of instances of event data comprises a user interaction with one of a feed or a notification, and the application system uses the model output to configure a notification portion of the user interface output. [0036] “user interactions with advertising, recommendations, content and each other in the social network,” and “[0023] “different types of content may therefore be served or recommended to be served, including advertisements, resumes of job seekers, job notifications, and/or others.”
Regarding Claim 17: Liu discloses: A method comprising:
by a feature generation system, recording a stream of user interface events into a first data store, wherein the first data store comprises a real-time data store; [0050-0051] (data pipeline to collect raw tracking data from user behaviors and interactions, i.e. record a stream of user interface events), [0105-0106] (obtain user interaction data), [0105] (stores events in memory 924), [0020] “the recommendations are computed and applied in real-time,” (such that recommendations computed and applied in real-time must imply real-time data storage);
obtaining, attributes that map to entities identified in particular events of the stream; [0109] (obtain attributes), [0108] (attributes map to user behavior events) [0079] extracts;
joining the attributes with corresponding events in the stream to produce processed event data; [0051] (raw tracking data), [0020] (tracking data is joined with attributed data), [0053] (information is used to provide behavioral and profile features);
receiving, from an application system that uses output of a machine learning model to respond to a user interface activity in the application system, a request for model output; [0107-0114] (API receives user activity features associated with user activities and user attributes and sends a request, received by the machine learning model, to make a recommendation used to configure a user interface in response to client device user activities);
sending, by the machine learning model, a feature request, to a feature generation system; [0072-0073]
reading, by the feature generation system, a feature configuration associated with a requested user activity feature; [0088]
determining, by the feature generation system, based on the feature configuration, a data access mechanism for a real-time data store a time window defined by the request timestamp, and a feature computation algorithm; [0062] (data is accessed according to user’s activities associated with interest segments and categorical content, i.e. the configuration of the features), [0059] ”time window,” and [0108] (different machine learning models containing feature appropriate algorithmic calculations for different feature configurations);
retrieving, from the real-time data store, the feature generation system using the data access mechanism, a plurality of instances of the processed event data that each comprise an event timestamp within the time window, and attribute data associated with the plurality of instances of event data; [0050] (data pipeline to collect dynamic actions), [0053] (plurality of instances of event data), [0059] (data collected according to “time windows”), [0020] “the recommendations are computed and applied in real-time,” (Examiner notes that real-time computations requires real-time data), [0079] (retrieve features, i.e. processed event data);
computing, by the feature generation system, the requested user activity feature using the plurality of instances of event data and the attribute data as inputs to the feature computation algorithm; [0053]
responsive to the feature request, providing, by the feature generation system, the computed user activity feature to the machine learning model; and [0053-0054]
responsive to the request for model output, by the machine learning model, generating a model output using the computed user activity feature as an input, and providing the model output to the application system. [0105-0108] (communications interface 112 receives requested machine learning model input) and [0114] (communications interface 112 sends model output).
Where Liu does not disclose Bhardwaj teaches:
and a request timestamp; [0015] “timestamp,”
a time window defined by the request timestamp, [0047] “event data may be grouped based on time windows (e.g., based on the timestamp of each event).”
an event timestamp within the time window; [0015]
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query databases and present social media and employment recommendations to user’s, utilizing collaborative filtering, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Where Liu does not disclose and Bhardwaj does not teach, LaBorde teaches:
by a write-side portion of a feature generation system, recording a stream of events [Column 10, Lines 59-68] (Command Query Responsibility Segregation, CQRS, where the command and query are segregated such that the stream of events is stored and may be reconstructed by the command side, and the query side is used to query for data to input into machine learning models; streams of user events are recorded in the command side, i.e. the command side is the write-side portion in the feature generation system);
by a read-side portion of the feature generation system; [Column 10, Lines 59-68] (Command Query Responsibility Segregation, CQRS, where the command and query sides are segregated such that the stream of events is received, stored, and may be reconstructed by the command side, i.e. the write-side, and the query side, i.e. the read-side, is used to query for data to input into machine learning models and return results), [Column 11, Lines 50-54] “providing data input into the system and the predictive analytic outputs of the proprietary models that consume this data. Data generated via these work flows executed in the client application in the regular course of its use;”
A first and second data store [Column 10, Lines 16-20, and Figure 4] (at least two different databases utilized for storage and retrieval of data in the CQRS, Command and Query portions of the system).
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query data collected in databases and present social media and employment recommendations to user’s, utilizing collaborative filtering and machine learning models, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Regarding Claim 18: Liu discloses: The method of claim 17, wherein computing the requested user activity feature comprises applying the feature computation algorithm to a portion of the plurality of instances of event data that matches a user identifier associated with the user interface activity. [0021]
Regarding Claim 19: Liu discloses: The method of claim 17, wherein computing the requested user activity feature comprises applying the feature computation algorithm to a portion of the plurality of instances of event data that matches an event identifier associated with the user interface activity. [0023]
Regarding Claim 20: Liu discloses: The method of claim 17, wherein computing the requested user activity feature comprises applying the feature computation algorithm to a portion of the plurality of instances of event data that matches a value of an attribute of one of an event associated with the user interface activity or an entity associated with the user interface activity. [0047]
Claim 5, 7, and 11-13 are rejected under 35 U.S.C. 103 as being unpatentable over as being unpatentable over Liu, US20150006295A1, in view of Bhardwaj, US20210035046A1, in further view of LaBorde, US10043591B1, and in further view of Gao, US20110093327A1.
Regarding Claim 5: Liu discloses, and Bhardwaj and LaBorde teach: The network-based service of claim 4,
Where Liu does not disclose and Bhardwaj and LaBorde do not teach: wherein the real-time data store is arranged according to a schema that is defined based on a feature type associated with the request.
Gao teaches: [0051].
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and statistical and machine learning methods, where the algorithms and machine learning models are programed to query databases and present social media and employment recommendations to user’s, utilizing collaborative filtering, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Regarding Claim 7: Liu discloses, and Bhardwaj and LaBorde teach: The network-based service of claim 1,
Where Liu does not disclose and Bhardwaj and LaBorde do not teach: wherein a maximum value of the time window is defined as N days prior to and including a day of the request timestamp and N is a positive integer
Gao teaches [0052] “Similarly, feature-level attributes 228 may include a data type representing the feature as a(n) integer,” [0061], [0068], [0069], and [0071] (Examiner notes that the visualization is described in [0068] as showing the maximum values of the feature using a time interval where [0061], [0069] and [0071] disclose the time interval may be manually set to any value.
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and machine learning for social media and employment networks with database querying to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system utilizing latency calculations.
Regarding Claim 11: Liu discloses, and Bhardwaj and LaBorde teach: The method of claim 10, wherein an instance of the plurality of instances of event data comprises one of a job search, a job view, a job application, or a job dismiss, and the application system uses the model output to configure a recommendation portion of the user interface output to include a recommendation to submit a job application for a particular job. [0023] “a job hunt (e.g., comprising applications or searches for jobs), “and [0087] “the one or more recommendations may include specific job opportunities that have been made known via the social network.”
Where Liu does not disclose and Bhardwaj and LaBorde do not teach: a recommendation to submit a job application for a particular job.
Gao teaches: [Fig. 6] “To apply click here.”
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and machine learning for social media and employment networks with database querying to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system utilizing latency calculations.
Regarding Claim 12: Liu discloses, and Bhardwaj and LaBorde teach: The method of claim 10,
Liu discloses: wherein an instance of the plurality of instances of event data comprises one of a profile view, a company view, a search query or a job application, and the application system uses the model output to configure a recommendation portion [0021] “The relevancy scores may be based on user behaviors such as online activities, searches, different types of page views, profile views, emails sent, etc.,” and [0023] (data features are extracted to configure recommendations)
Where Liu does not disclose and Bhardwaj and LaBorde do not teach: a recommendation portion of the user interface output to include a recommendation to send a connection request to a particular other user of the application system.
Gao teaches: [0021] “and/or recommend one another,” [0025] “may include features or mechanisms for recommending connections, job postings, articles, and/or groups to the entities.”
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and machine learning, where the algorithms and machine learning models are programed to query databases and present recommendations to a user, utilizing collaborative filtering, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Regarding Claim 13: Liu discloses, and Bhardwaj and LaBorde teach: The method of claim 10,
Liu discloses: the application system uses the model output to configure a recommendation portion wherein an instance of the plurality of instances of event data comprises a user interaction with one of a feed or a post, [0021] “The relevancy scores may be based on user behaviors such as online activities, searches, different types of page views, profile views, emails sent, etc.,” and [0023] (data features are extracted to configure recommendations)
Where Liu does not disclose and Bhardwaj and LaBorde do not teach: to configure a recommendation portion of the user interface output to include a recommendation to follow one of a particular user of the application system or a particular topic in the application system
Gao teaches: [0021] “and/or recommend one another,” [0025] “may include features or mechanisms for recommending connections, job postings, articles, and/or groups to the entities.”
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and machine learning, where the algorithms and machine learning models are programed to query databases and present recommendations to a user, utilizing collaborative filtering, to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of social media and employment recommenders.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Liu, US20150006295A1, in view of Bhardwaj, US20210035046A1, and in further view of LaBorde, US10043591B1, in further view of Soman, “Achieving 99th Percentile Latency SLA Using Apache Pinot,” 10/1/2021. (Examiner notes that the instant application discloses Apache Pinot and Kafka in the disclosure, while LinkedIn utilizing the time window latency of 100ms is disclosed in the prior art).
Regarding Claim 8: Liu discloses, and Bhardwaj and LaBorde teach: The network-based service of claim 1,
173. Where Liu does not disclose and Bhardwaj and LaBorde do not teach: wherein a difference between the request timestamp and a timestamp at which the user activity feature is computed is less than 100 milliseconds.
Soman teaches: “Let’s take LinkedIn’s Feed ranking use case which leverages Pinot for curating the home page feed recommendations to its 700+ million users. For a given feed item, we want to know how many times has a given user seen it to decide whether it should be displayed. This can be done using the following query: SELECT sum(count) from T WHERE memberId = X AND item in (list of 500-1500 items) AND time >= (now - 14 days) GROUP BY action, item, position, time. This will run for every single active user visiting the LinkedIn home page, translating to multiple 1000s of OLAP queries to Pinot. For a good user experience, it is critical that the p99th latency is under 100ms. This is non-trivial to achieve since traditional RDBMS/OLTP stores don’t scale for such analytical workloads. Even though Apache Pinot was built for such use cases, we need to tune it correctly to ensure p99th SLA.”
One of ordinary skill in the art would find it obvious to combine the prior art before the effective filing date because they share the same field of computer science in algorithms and machine learning, and assignee, LinkedIn, with low latency social network modeling and querying to yield the predictable results of the combined disclosure and patent claims. The combination of these prior art documents leads to a markedly advantageous system of low latency timeframes.
Conclusion
Examiner note: depending on the output to the user interface, the 5 U.S.C. § 101 analysis could also represent abstract ideas in methods of organizing human activity in commercial or legal interactions as they also drive business relations and employment contracts when the system is used in relation to at least employment actions.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANGELA HATCH whose telephone number is (571)270-1393. The examiner can normally be reached 10:00-6:00.
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, Nathan Uber can be reached at (571)270-3923. 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.
ANGELA HATCH
Examiner
Art Unit 3626
/ANGELA HATCH/Examiner, Art Unit 3626
/NATHAN C UBER/Supervisory Patent Examiner, Art Unit 3626
Prosecution Insights