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 . Claims 1-20 are presented in the case.
Priority
Applicant's claim for the benefit of a prior-filed Provisional application 63/469,854 filed on 04/18/2023 is acknowledged.
Information Disclosure Statement
The information disclosure statement submitted on 09/27/2024 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The term “insufficient” listed twice in claim 1 is a relative term which renders the claim indefinite. The term “insufficient” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. Is this a model not a model yet or does this need to meet some sort of accuracy level before raw data or training is sufficient?
The term clean is used in conjunction with “the raw data set” and “raw data” it is uncertain which data is being cleaned in claim 1. Then extract history from raw data that is cleaned is unclear. Should this be the raw data set?
The term “transaction” lacks a clear antecedent and the relationship between fraud event and transaction is unclear in claim 5, 12 and 19.
The term “valid” is a relative term in claim 5, 12 and 19.
The term “shared user accounts” should be “shared user account” in claim 11 and 18.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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 of this title, 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.
Claims 1-4, 6-11, 13-18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Meng et al. (US 20200045066 A1) hereinafter Meng in view of Abreu et al. (US 20220027916 A1) hereinafter Abreu and Turgeman et al. (US 20140325646 A1) hereinafter Turgeman.
As to independent claim 1, Meng teaches a system for creating labels and training a machine learning model using a limited set of data, the system comprising: [system that generates labels and trains ¶11-12]
a client device including a first electronic processor and a first memory; [client ¶78; computer with processor and memory ¶74]
a user device including a second electronic processor and a second memory; [user computer ¶77; computer with processor and memory ¶74]
a storage device including a third electronic processor and a third memory, the storage device associated with the client device and the user device; and [data server ¶77; computer with processor and memory ¶74]
a server including a fourth electronic processor and a fourth memory including a label creation application, the fourth electronic processor configured to: [server ¶78; computer with processor and memory ¶74]
receive raw data from the storage device, wherein the raw data includes a plurality of requests associated with a plurality of user accounts, the plurality of user accounts associated with the user device, [stores application events (raw data from requests) related to accounts ¶36, ¶52 "account sign up, login, online transactions ", ¶33 "continuously receives and stores application-level events 12, generated by users, from multiple online service providers 14."]wherein the raw data is insufficient for training the machine learning model, [overcomes lack of labeled data (insufficient) with augmenting auto labels ¶11, ¶14, ¶32 "overcome the lack of labeled data in security and/or fraud applications, the deep learning system can make use of automatically-generated labels"]
determine, with the label creation application, an account type of each of the plurality of user accounts, [predicts if account is fraudulent or not (account type) ¶10 "predict, given a set of anonymized common digital information in an application-level event, whether the user account associated with the event is fraudulent"]
generate, with the label creation application using the raw data, a raw data set based on [[the account type that is determined,]] the plurality of requests, and the plurality of user accounts, [generates information based on users, requests and accounts (user sign-up, transactions etc.) ¶33-40, ¶50 "sign up, login, online transactions (e.g., for financial services or e-commerce services), post, like, follow, message (e.g., for social services), or other application-specific actions"].
wherein the limited set of data is insufficient for training the machine learning model, [overcomes lack of labeled data (insufficient) with augmenting auto labels ¶11, ¶14, ¶32 "overcome the lack of labeled data in security and/or fraud applications, the deep learning system can make use of automatically-generated labels"]
extract, with the label creation application, a request history for a user account from the raw data that is cleaned, [derives user behavior and uses past actions (history) ¶16, 36-40 "actions the respective individual users perform when interacting with a particular online service provider"]
generate, with the label creation application, a training profile associated with the user account of the raw data based on the request history that is extracted, [generates training vectors ¶39-40 "generate a feature vector for each user and/or each application-level event,"]
create, with the label creation application, training labels for training the machine learning model based on the training profile associated with the user account of the raw data that is cleaned, and [generates labels ¶11-12, 50, 55-59 " training labels are generated by unsupervised machine learning fraud detection algorithms"]
process, with the machine learning model, the training profile and the training labels that are created to train the machine learning model. [trains accordingly ¶34, 50-51, 59 "trains a deep learning neural network model 22 using the input data."]
Meng does not specifically clean, with the label creation application, the raw data set using client feedback data from the client device, wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client.
However, Abreu teaches clean, with the label creation application, the raw data set using client feedback data from the client device, [cleanse data using feedback data including fraud events ¶51, ¶55-58 "cleansing module 310 operates to clean and standardize the received feedback data. This may include, for example, renaming labels associated with data"]
wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client, [reported and known fraud data with transaction data ¶37, ¶62 "recently reported fraudulent transactions"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify training data disclosed by Meng by incorporating the clean, with the label creation application, the raw data set using client feedback data from the client device, wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client by Abreu because both techniques address the same field of machine learning and by incorporating Abreu into Meng helps reduce online fraud with more ideal models that more accurately identifies fraud in transactions [Abreu ¶5]
Meng and Abreu do not specifically teach generates raw data based on account type.
However, Turgeman teaches generates raw data based on account type [detects if accounts are associated with one or multiple users (joint or single account type) and then uses behavior data in training ¶102 "learns the user behavior and builds a model", ¶112-113, ¶96 " take into consideration also the fact that this bank account is jointly-owned by a married couple of two senior citizens; thereby allowing the second access session without raising a possible fraud alert."].
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the training data disclosed by Meng and Abreu by incorporating the generates raw data based on account type disclosed by Turgeman because all techniques address the same field of fraud detection and by incorporating Turgeman into Meng and Abreu better identifies fraudulent users reducing attacks on the systems [Turgeman ¶9-10]
As to dependent claim 2, the rejection of claim 1 is incorporated. Meng, Abreu and Turgeman further teach wherein determining the account type of each of the plurality of user accounts, the fourth electronic processor is further configured to:
determine a first user account of the plurality of user accounts is associated with a single user based on a set of heuristic conditions associated with identifying accounts used by a single user, and [Turgeman detects credentials (heuristic) based on time (conditions) ¶112 " detect that two (or more) different users are accessing, or are attempting to access, at different times or during overlapping or partially-overlapping time-periods, the same computerized service, using the same user-account"]
assign the first user account of the plurality of user accounts a first account type that is associated with single user accounts. [Turgeman detects if account is used by one user or not and acts accordingly ¶98-99, ¶101]
As to dependent claim 3, the rejection of claim 2 is incorporated. Meng, Abreu and Turgeman further teach wherein creating training labels for training the machine learning model, the fourth electronic processor is further configured to:
determine a device type associated with the user account, and [Turgeman hardware type ¶70, ¶51 "determine the type of hardware utilized by a user, and thereby assist in distinguishing between a local user versus a remote attacker."]
create a request associated with the user account by replacing a first device type associated with a first request with a second device type associated with the user account, and [Turgeman detects then changes request according to touch vs mouse vs keyboard use (different types of devices) ¶70-80 "a genuine user may be located in the United States and may utilize an American English keyboard"]
assign a label to the request that is created. [Meng feedback requests add labels ¶35, ¶57-59 "feedback on specific instances provides labeled data that can be used to tune the parameters of the model"]
As to dependent claim 4, the rejection of claim 1 is incorporated. Meng, Abreu and Turgeman further teach wherein determining the account type of each of the plurality of user accounts, the fourth electronic processor is further configured to:
determine a second user account of the plurality of user accounts is associated with two or more users based on a set of heuristic conditions associated with identifying accounts shared by users, and [Turgeman detects credentials (heuristic) based on time (conditions) ¶112 " detect that two (or more) different users are accessing, or are attempting to access, at different times or during overlapping or partially-overlapping time-periods, the same computerized service, using the same user-account"]
assign the second user account a second account type that is associated with a shared user account. [Turgeman detects if account is used by one user or not and acts accordingly ¶98-99, ¶101]
As to dependent claim 6, the rejection of claim 4 is incorporated. Meng, Abreu and Turgeman further teach wherein creating training labels for training the machine learning model, the fourth electronic processor is further configured to:
create a request associated with a first user of the two or more users by replacing a first request associated with an identifier of the first user with a second request associated with an identifier of a second user of the two or more users, and [Turgeman detects fraud and sends request that replaces or augments with username (identifier) ¶65 "The malware may then capture the additional data that the user enters and/or submits, while transmitting back to the web-server only the data for the originally-required fields (the username and the password) and not the augmented (fraudulent) fields."]
assign a label to the request that is created. [Meng feedback requests add labels ¶35, ¶57-59 "feedback on specific instances provides labeled data that can be used to tune the parameters of the model"]
As to dependent claim 7, the rejection of claim 4 is incorporated. Meng, Abreu and Turgeman further teach wherein generating the raw data set, the fourth electronic processor is further configured to:
filter the raw data using a set of heuristic conditions, and [Turgeman detects users using credentials and conditions of time, clicks etc. ¶112, ¶117 filters down to user specific training data ¶102 "utilize 5-10 sessions per user (not per account) to build the model"]
create a subset of the raw data using the raw data that satisfies the set of heuristic conditions. [Turgeman 5-10 session data is a subset ¶102]
As to independent claim 8, Meng teaches a method for creating labels for training a machine learning model using a limited set of data, the method comprising: [generates labels and trains ¶11-12]
a client device including a first electronic processor and a first memory; [client ¶78; computer with processor and memory ¶74]
a user device including a second electronic processor and a second memory; [user computer ¶77; computer with processor and memory ¶74]
a storage device including a third electronic processor and a third memory, the storage device associated with the client device and the user device; and [data server ¶77; computer with processor and memory ¶74]
a server including a fourth electronic processor and a fourth memory including a label creation application, the fourth electronic processor configured to: [server ¶78; computer with processor and memory ¶74]
receive raw data from the storage device, wherein the raw data includes a plurality of requests associated with a plurality of user accounts, the plurality of user accounts associated with the user device, [stores application events (raw data from requests) related to accounts ¶36, ¶52 "account sign up, login, online transactions ", ¶33 "continuously receives and stores application-level events 12, generated by users, from multiple online service providers 14."]wherein the raw data is insufficient for training the machine learning model, [overcomes lack of labeled data (insufficient) with augmenting auto labels ¶11, ¶14, ¶32 "overcome the lack of labeled data in security and/or fraud applications, the deep learning system can make use of automatically-generated labels"]
determine, with the label creation application, an account type of each of the plurality of user accounts, [predicts if account is fraudulent or not (account type) ¶10 "predict, given a set of anonymized common digital information in an application-level event, whether the user account associated with the event is fraudulent"]
generate, with the label creation application using the raw data, a raw data set based on [[the account type that is determined,]] the plurality of requests, and the plurality of user accounts, [generates information based on users, requests and accounts (user sign-up, transactions etc.) ¶33-40, ¶50 "sign up, login, online transactions (e.g., for financial services or e-commerce services), post, like, follow, message (e.g., for social services), or other application-specific actions"].
wherein the limited set of data is insufficient for training the machine learning model, [overcomes lack of labeled data (insufficient) with augmenting auto labels ¶11, ¶14, ¶32 "overcome the lack of labeled data in security and/or fraud applications, the deep learning system can make use of automatically-generated labels"]
extract, with the label creation application, a request history for a user account from the raw data that is cleaned, [derives user behavior and uses past actions (history) ¶16, 36-40 "actions the respective individual users perform when interacting with a particular online service provider"]
generate, with the label creation application, a training profile associated with the user account of the raw data based on the request history that is extracted, [generates training vectors ¶39-40 "generate a feature vector for each user and/or each application-level event,"]
create, with the label creation application, training labels for training the machine learning model based on the training profile associated with the user account of the raw data that is cleaned, and [generates labels ¶11-12, 50, 55-59 " training labels are generated by unsupervised machine learning fraud detection algorithms"]
process, with the machine learning model, the training profile and the training labels that are created to train the machine learning model. [trains accordingly ¶34, 50-51, 59 "trains a deep learning neural network model 22 using the input data."]
Meng does not specifically clean, with the label creation application, the raw data set using client feedback data from the client device, wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client.
However, Abreu teaches clean, with the label creation application, the raw data set using client feedback data from the client device, [cleanse data using feedback data including fraud events ¶51, ¶55-58 "cleansing module 310 operates to clean and standardize the received feedback data. This may include, for example, renaming labels associated with data"]
wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client, [reported and known fraud data with transaction data ¶37, ¶62 "recently reported fraudulent transactions"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify training data disclosed by Meng by incorporating the clean, with the label creation application, the raw data set using client feedback data from the client device, wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client by Abreu because both techniques address the same field of machine learning and by incorporating Abreu into Meng helps reduce online fraud with more ideal models that more accurately identifies fraud in transactions [Abreu ¶5]
Meng and Abreu do not specifically teach generates raw data based on account type.
However, Turgeman teaches generates raw data based on account type [detects if accounts are associated with one or multiple users (joint or single account type) and then uses behavior data in training ¶102 "learns the user behavior and builds a model", ¶112-113, ¶96 " take into consideration also the fact that this bank account is jointly-owned by a married couple of two senior citizens; thereby allowing the second access session without raising a possible fraud alert."].
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the training data disclosed by Meng and Abreu by incorporating the generates raw data based on account type disclosed by Turgeman because all techniques address the same field of fraud detection and by incorporating Turgeman into Meng and Abreu better identifies fraudulent users reducing attacks on the systems [Turgeman ¶9-10]
As to dependent claim 9, the rejection of claim 8 is incorporated. Meng, Abreu and Turgeman further teach wherein determining the account type of each of the plurality of user accounts, the method further comprises:
determining a first user account of the plurality of user accounts is associated with a single user based on a set of heuristic conditions associated with identifying accounts used by a single user, and [Turgeman detects credentials (heuristic) based on time (conditions) ¶112 " detect that two (or more) different users are accessing, or are attempting to access, at different times or during overlapping or partially-overlapping time-periods, the same computerized service, using the same user-account"]
assigning the first user account of the plurality of user accounts a first account type that is associated with single user accounts. [Turgeman detects if account is used by one user or not and acts accordingly ¶98-99, ¶101]
As to dependent claim 10, the rejection of claim 9 is incorporated. Meng, Abreu and Turgeman further teach wherein creating training labels for training the machine learning model, the method further comprises:
determining a device type associated with the user account, and [Turgeman hardware type ¶70, ¶51 "determine the type of hardware utilized by a user, and thereby assist in distinguishing between a local user versus a remote attacker."]
creating a request associated with the user account by replacing a first device type associated with a first request with a second device type associated with the user account, and [Turgeman detects then changes request according to touch vs mouse vs keyboard use (different types of devices) ¶70-80 "a genuine user may be located in the United States and may utilize an American English keyboard"]
assigning a label to the request that is created. [Meng feedback requests add labels ¶35, ¶57-59 "feedback on specific instances provides labeled data that can be used to tune the parameters of the model"]
As to dependent claim 11, the rejection of claim 8 is incorporated. Meng, Abreu and Turgeman further teach wherein determining the account type of each of the plurality of user accounts, the method further comprises:
determining a second user account of the plurality of user accounts is associated with two or more users based on a set of heuristic conditions associated with identifying accounts shared by users, and [Turgeman detects credentials (heuristic) based on time (conditions) ¶112 " detect that two (or more) different users are accessing, or are attempting to access, at different times or during overlapping or partially-overlapping time-periods, the same computerized service, using the same user-account"]
assigning the second user account a second account type that is associated with a shared user account. [Turgeman detects if account is used by one user or not and acts accordingly ¶98-99, ¶101]
As to dependent claim 13, the rejection of claim 11 is incorporated. Meng, Abreu and Turgeman further teach wherein creating training labels for training the machine learning model, the fourth electronic processor is further configured to:
creating a request associated with a first user of the two or more users by replacing a first request associated with an identifier of the first user with a second request associated with an identifier of a second user of the two or more users, and [Turgeman detects fraud and sends request that replaces or augments with username (identifier) ¶65 "The malware may then capture the additional data that the user enters and/or submits, while transmitting back to the web-server only the data for the originally-required fields (the username and the password) and not the augmented (fraudulent) fields."]
assigning a label to the request that is created. [Meng feedback requests add labels ¶35, ¶57-59 "feedback on specific instances provides labeled data that can be used to tune the parameters of the model"]
As to dependent claim 14, the rejection of claim 8 is incorporated. Meng, Abreu and Turgeman further teach wherein generating the raw data set, the method further comprises:
filtering the raw data using a set of heuristic conditions, and [Turgeman detects users using credentials and conditions of time, clicks etc. ¶112, ¶117 filters down to user specific training data ¶102 "utilize 5-10 sessions per user (not per account) to build the model"]
creating a subset of the raw data using the raw data that satisfies the set of heuristic conditions. [Turgeman 5-10 session data is a subset ¶102]
As to independent claim 15, Meng teaches a method for creating labels for training a machine learning model using a limited set of data, the method comprising: [generates labels and trains ¶11-12]
a client device including a first electronic processor and a first memory; [client ¶78; computer with processor and memory ¶74]
a user device including a second electronic processor and a second memory; [user computer ¶77; computer with processor and memory ¶74]
a storage device including a third electronic processor and a third memory, the storage device associated with the client device and the user device; and [data server ¶77; computer with processor and memory ¶74]
a server including a fourth electronic processor and a fourth memory including a label creation application, the fourth electronic processor configured to: [server ¶78; computer with processor and memory ¶74]
receive raw data from the storage device, wherein the raw data includes a plurality of requests associated with a plurality of user accounts, the plurality of user accounts associated with the user device, [stores application events (raw data from requests) related to accounts ¶36, ¶52 "account sign up, login, online transactions ", ¶33 "continuously receives and stores application-level events 12, generated by users, from multiple online service providers 14."]wherein the raw data is insufficient for training the machine learning model, [overcomes lack of labeled data (insufficient) with augmenting auto labels ¶11, ¶14, ¶32 "overcome the lack of labeled data in security and/or fraud applications, the deep learning system can make use of automatically-generated labels"]
determine, with the label creation application, an account type of each of the plurality of user accounts, [predicts if account is fraudulent or not (account type) ¶10 "predict, given a set of anonymized common digital information in an application-level event, whether the user account associated with the event is fraudulent"]
generate, with the label creation application using the raw data, a raw data set based on [[the account type that is determined,]] the plurality of requests, and the plurality of user accounts, [generates information based on users, requests and accounts (user sign-up, transactions etc.) ¶33-40, ¶50 "sign up, login, online transactions (e.g., for financial services or e-commerce services), post, like, follow, message (e.g., for social services), or other application-specific actions"].
wherein the limited set of data is insufficient for training the machine learning model, [overcomes lack of labeled data (insufficient) with augmenting auto labels ¶11, ¶14, ¶32 "overcome the lack of labeled data in security and/or fraud applications, the deep learning system can make use of automatically-generated labels"]
extract, with the label creation application, a request history for a user account from the raw data that is cleaned, [derives user behavior and uses past actions (history) ¶16, 36-40 "actions the respective individual users perform when interacting with a particular online service provider"]
generate, with the label creation application, a training profile associated with the user account of the raw data based on the request history that is extracted, [generates training vectors ¶39-40 "generate a feature vector for each user and/or each application-level event,"]
create, with the label creation application, training labels for training the machine learning model based on the training profile associated with the user account of the raw data that is cleaned, and [generates labels ¶11-12, 50, 55-59 " training labels are generated by unsupervised machine learning fraud detection algorithms"]
process, with the machine learning model, the training profile and the training labels that are created to train the machine learning model. [trains accordingly ¶34, 50-51, 59 "trains a deep learning neural network model 22 using the input data."]
Meng does not specifically clean, with the label creation application, the raw data set using client feedback data from the client device, wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client.
However, Abreu teaches clean, with the label creation application, the raw data set using client feedback data from the client device, [cleanse data using feedback data including fraud events ¶51, ¶55-58 "cleansing module 310 operates to clean and standardize the received feedback data. This may include, for example, renaming labels associated with data"]
wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client, [reported and known fraud data with transaction data ¶37, ¶62 "recently reported fraudulent transactions"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify training data disclosed by Meng by incorporating the clean, with the label creation application, the raw data set using client feedback data from the client device, wherein the client feedback data is the limited set of data that includes one or more fraud events associated with one or more user accounts and identified by a client by Abreu because both techniques address the same field of machine learning and by incorporating Abreu into Meng helps reduce online fraud with more ideal models that more accurately identifies fraud in transactions [Abreu ¶5]
Meng and Abreu do not specifically teach generates raw data based on account type.
However, Turgeman teaches generates raw data based on account type [detects if accounts are associated with one or multiple users (joint or single account type) and then uses behavior data in training ¶102 "learns the user behavior and builds a model", ¶112-113, ¶96 " take into consideration also the fact that this bank account is jointly-owned by a married couple of two senior citizens; thereby allowing the second access session without raising a possible fraud alert."].
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the training data disclosed by Meng and Abreu by incorporating the generates raw data based on account type disclosed by Turgeman because all techniques address the same field of fraud detection and by incorporating Turgeman into Meng and Abreu better identifies fraudulent users reducing attacks on the systems [Turgeman ¶9-10]
As to dependent claim 16, the rejection of claim 15 is incorporated. Meng, Abreu and Turgeman further teach wherein determining the account type of each of the plurality of user accounts, further comprises:
determining a first user account of the plurality of user accounts is associated with a single user based on a set of heuristic conditions associated with identifying accounts used by a single user, and [Turgeman detects credentials (heuristic) based on time (conditions) ¶112 " detect that two (or more) different users are accessing, or are attempting to access, at different times or during overlapping or partially-overlapping time-periods, the same computerized service, using the same user-account"]
assigning the first user account of the plurality of user accounts a first account type that is associated with single user accounts. [Turgeman detects if account is used by one user or not and acts accordingly ¶98-99, ¶101]
As to dependent claim 17, the rejection of claim 16 is incorporated. Meng, Abreu and Turgeman further teach wherein creating training labels for training the machine learning model, further comprises:
determining a device type associated with the user account, and [Turgeman hardware type ¶70, ¶51 "determine the type of hardware utilized by a user, and thereby assist in distinguishing between a local user versus a remote attacker."]
creating a request associated with the user account by replacing a first device type associated with a first request with a second device type associated with the user account, and [Turgeman detects then changes request according to touch vs mouse vs keyboard use (different types of devices) ¶70-80 "a genuine user may be located in the United States and may utilize an American English keyboard"]
assigning a label to the request that is created. [Meng feedback requests add labels ¶35, ¶57-59 "feedback on specific instances provides labeled data that can be used to tune the parameters of the model"]
As to dependent claim 18, the rejection of claim 15 is incorporated. Meng, Abreu and Turgeman further teach wherein determining the account type of each of the plurality of user accounts, further comprises:
determining a second user account of the plurality of user accounts is associated with two or more users based on a set of heuristic conditions associated with identifying accounts shared by users, and [Turgeman detects credentials (heuristic) based on time (conditions) ¶112 " detect that two (or more) different users are accessing, or are attempting to access, at different times or during overlapping or partially-overlapping time-periods, the same computerized service, using the same user-account"]
assigning the second user account a second account type that is associated with a shared user account. [Turgeman detects if account is used by one user or not and acts accordingly ¶98-99, ¶101]
As to dependent claim 20, the rejection of claim 18 is incorporated. Meng, Abreu and Turgeman further teach wherein creating training labels for training the machine learning model, further comprises:
creating a request associated with a first user of the two or more users by replacing a first request associated with an identifier of the first user with a second request associated with an identifier of a second user of the two or more users, and [Turgeman detects fraud and sends request that replaces or augments with username (identifier) ¶65 "The malware may then capture the additional data that the user enters and/or submits, while transmitting back to the web-server only the data for the originally-required fields (the username and the password) and not the augmented (fraudulent) fields."]
assigning a label to the request that is created. [Meng feedback requests add labels ¶35, ¶57-59 "feedback on specific instances provides labeled data that can be used to tune the parameters of the model"]
Claims 5, 12 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Meng in view of Abreu and Turgeman, as applied in claim 4, 11 and 18 above, and further in view of Hartmann (US 7690037 B1).
As to dependent claim 5, Meng, Abreu and Turgeman teach the method of claim 1 above that is incorporated,
Meng, Abreu and Turgeman further teach identify, using the client feedback data, a third user account of the one or more user accounts that is associated with an occurrence of a fraud event of the one or more fraud events, and [Meng feedback used to determine status of user account regarding fraud prediction made from events ¶58-59 " receives feedback from different client service providers regarding the accuracy of respective predictions 304. For example, based on a given prediction, a service provider can perform further investigation into the user account. In some cases the prediction is confirmed. In other cases, a false positive or false negative result may be uncovered. The service provider can provide the system with information on the actual status of the user account and whether that agreed or disagreed with the system prediction"]
Meng, Abreu and Turgeman do not specifically teach wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: remove the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model.
However, Hartmann teaches wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: remove the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model. [remove training data that is illegitimate or anomalous (fraud) Col. 1-2 ln. 56-6 " data from the clusters representing actual anomalous activities are excluded from the corpus. As a result, machine learning based on the corpus is more effective and the trained system provides better performance, since latent anomalies are not mistaken for normal activity"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the account management disclosed by Meng, Abreu and Turgeman by incorporating wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: remove the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model disclosed by Hartmann because all techniques address the same field of machine learning and by incorporating Hartmann into Meng, Abreu and Turgeman generates less anomalous activities allowing more effective training for better performance [Hartmann Col. 1 ln. 50-6].
As to dependent claim 12, Meng, Abreu and Turgeman teach the method of claim 11 above that is incorporated,
Meng, Abreu and Turgeman further teach identifying, using the client feedback data, a third user account of the one or more user accounts that is associated with an occurrence of a fraud event of the one or more fraud events, and [Meng feedback used to determine status of user account regarding fraud prediction made from events ¶58-59 " receives feedback from different client service providers regarding the accuracy of respective predictions 304. For example, based on a given prediction, a service provider can perform further investigation into the user account. In some cases the prediction is confirmed. In other cases, a false positive or false negative result may be uncovered. The service provider can provide the system with information on the actual status of the user account and whether that agreed or disagreed with the system prediction"]
Meng, Abreu and Turgeman do not specifically teach wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: removing the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model.
However, Hartmann teaches wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: removing the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model. [remove training data that is illegitimate or anomalous (fraud) Col. 1-2 ln. 56-6 " data from the clusters representing actual anomalous activities are excluded from the corpus. As a result, machine learning based on the corpus is more effective and the trained system provides better performance, since latent anomalies are not mistaken for normal activity"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the account management disclosed by Meng, Abreu and Turgeman by incorporating wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: removing the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model disclosed by Hartmann because all techniques address the same field of machine learning and by incorporating Hartmann into Meng, Abreu and Turgeman generates less anomalous activities allowing more effective training for better performance [Hartmann Col. 1 ln. 50-6].
As to dependent claim 19, Meng, Abreu and Turgeman teach the method of claim 18 above that is incorporated,
Meng, Abreu and Turgeman further teach identifying, using the client feedback data, a third user account of the one or more user accounts that is associated with an occurrence of a fraud event of the one or more fraud events, and [Meng feedback used to determine status of user account regarding fraud prediction made from events ¶58-59 " receives feedback from different client service providers regarding the accuracy of respective predictions 304. For example, based on a given prediction, a service provider can perform further investigation into the user account. In some cases the prediction is confirmed. In other cases, a false positive or false negative result may be uncovered. The service provider can provide the system with information on the actual status of the user account and whether that agreed or disagreed with the system prediction"]
Meng, Abreu and Turgeman do not specifically teach wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: removing the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model.
However, Hartmann teaches wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: removing the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model. [remove training data that is illegitimate or anomalous (fraud) Col. 1-2 ln. 56-6 " data from the clusters representing actual anomalous activities are excluded from the corpus. As a result, machine learning based on the corpus is more effective and the trained system provides better performance, since latent anomalies are not mistaken for normal activity"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the account management disclosed by Meng, Abreu and Turgeman by incorporating wherein cleaning the raw data set using client feedback data, the fourth electronic processor is further configured to: removing the third user account of the one or more user accounts from the raw data set based on the fraud event and transaction of the third user account, wherein the third user account is not valid for training the machine learning model disclosed by Hartmann because all techniques address the same field of machine learning and by incorporating Hartmann into Meng, Abreu and Turgeman generates less anomalous activities allowing more effective training for better performance [Hartmann Col. 1 ln. 50-6].
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 C.F.R. § 1.111(c) to consider these references fully when responding to this action.
BIAGI et al. (US 20240330933 A1) teaches fraud labels as training data for classification models (see ¶25).
It is noted that any citation to specific pages, columns, lines, or figures in the prior art references and any interpretation of the references should not be considered to be limiting in any way. A reference is relevant for all it contains and may be relied upon for all that it would have reasonably suggested to one having ordinary skill in the art. In re Heck, 699 F.2d 1331, 1332-33, 216 U.S.P.Q. 1038, 1039 (Fed. Cir. 1983) (quoting In re Lemelson, 397 F.2d 1006, 1009, 158 U.S.P.Q. 275, 277 (C.C.P.A. 1968)).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Beau Spratt whose telephone number is 571 272 9919. The examiner can normally be reached 8:30am to 5:00pm (EST).
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, Jennifer Welch can be reached at 571 272 7212. The fax phone number for the organization where this application or proceeding is assigned is 571 483 7388.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866 217 9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800 786 9199 (IN USA OR CANADA) or 571 272 1000.
/BEAU D SPRATT/ Primary Examiner, Art Unit 2143
Prosecution Insights