DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This final office action is in response to the amendment filed 21 August 2026.
Claims 1-20 are pending. Claims 1, 12, and 20 are independent claims.
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 (i.e., changing from AIA to pre-AIA ) 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, 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Domenikos et al. (US 2008/0103798, published 1 May 2008, hereafter Domenikos) and further in view of Lightbody et al. (US 11386456, filed 3 December 2020, hereafter Lightbody) and further in view of Cox et al. (US 2017/0286622, published 5 October 2017, hereafter Cox) and further in view of Meyer et al. (US 2019/0378619, published 12 December 2019, hereafter Meyer) and further in view of Cifarelli et al. (US 2022/0188827, filed 10 December 2020, hereafter Cifarelli).
As per independent claim 1, Domenikos discloses a computing platform comprising:
at least one processor (paragraph 0111)
a communication interface communicatively coupled to the at least one processor (paragraph 0111)
memory storing computer-readable instructions that, when executed by the at least one processor (paragraph 0111), cause the computing platform to:
train, using neural network regression, a predictive identity scoring model configured to predict identity threat events (Figure 19; paragraph 0270).
receive new event information (paragraph 0065)
input the new event information into the machine learning model, wherein inputting the new event information into the machine learning model causes the machine learning model to output the identity information (Figure 5; paragraph 0071)
send, to a client device, the identity information and one or more commands directing the client device to display an identity information, wherein sending the identity information and one or more commands directing the client device to display the identity interface causes the client device to display the identity interface (Figure 8; paragraph 0090)
Domenikos fails to specifically disclose:
(I) create a selected time static classifier model based on historical event information, wherein creating the selected time static classifier model comprises:
(i) classifying the historical event information, wherein classifying the historical event information comprises:
(a) training the selected time static classifier model based on the historical event information
(1) creating, from the historical event information, positive and negative cases for training the selected time classifier model
(2) training a selected time static classifier model such that a time static classifier is trained to distinguish between positive and negative cases in the historical event information separately from time series data
wherein a user remediation occurrence of a fraud alert indicative of identity theft is noted such that the fraud alert is flagged as a positive case for a user, and wherein no user remediation occurrence of the fraud alert indicative of identity theft is noted such that the fraud alert is flagged as a negative case for the user
(II) after classifying the historical event information, performing time series calibration on the classified historical event information, wherein training the machine learning model configures the machine learning model to output identity information
wherein performing time series calibration comprises inserting the trained time static classifier into the time series data such that the time static classifier is integrated into the time series data
However, Lightbody, which is analogous to the claimed invention because it is directed toward generating training datasets, discloses:
(I) create a selected time static classifier model based on historical event information (column 4, line 60- column 5, line 11: Here, a training dataset is generated for time series data (column 5, lines 12-37). This includes labeling historical data with labels to create a ground truth dataset for training a model)
wherein creating the selected time static classifier model comprises: (i) classifying the historical event information (column 4, line 60- column 5, line 11: Here, a training dataset is generated for time series data (column 5, lines 12-37). This includes labeling historical data with labels to create a ground truth dataset for training a model)
wherein classifying the historical event information comprises: (a) training the selected time static classifier model based on the historical event information (column 4, line 60- column 5, line 11: Here, a training dataset is generated for time series data (column 5, lines 12-37). This includes labeling historical data with labels to create a ground truth dataset for training a model)
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Lightbody with Domenikos, with a reasonable expectation of success, as it would have allowed for generating ground truth datasets for training a model (Lightbody: column 4, line 60- column 5, line 11).
However, Cox, which is analogous to the claimed invention because it is directed toward using machine learning to perform a patient risk assessment, discloses:
(II) after classifying the historical event information, performing time series calibration on the classified historical event information, wherein training the machine learning model configures the machine learning model to output identity information (Figure 16; paragraph 0246)
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s expected filing date to have combined Cox with Domenikos-Lightbody, with a reasonable expectation of success, as it would have allowed for updating a patient risk assessment via a machine learning model based upon new information. This would have provided the user with flexible and updatable analysis of a user’s data.
However, Meyer, which is analogous to the claimed invention because it is directed toward using machine learning to predict conditions, discloses:
wherein classifying the historical even information comprises: training a selected time static classifier model such that a time static classifier is trained to distinguish between positive and negative cases in the historical event information separately from time series data (paragraphs 0004-0005 and 0020: Here, a first set of static data comprising features of data and a second set of dynamic data comprising features of data are received. The data is processed absent time data and assigning the values for the features to predict the conditions of the patient and time series data based on the first and second set)
wherein performing time series calibration comprises inserting the trained time static classifier into the time series data such that the time static classifier is integrated into the time series data (paragraphs 0006 and 0020: Here, the assigned values for features and associated time series data are provided to a trained machine learning model to predict the likelihood of a condition to occur at a future time)
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Meyer with Domenikos-Lightbody-Cox, with a reasonable expectation of success, as it would have allowed for improved training and prediction by a neural network for identifying conditions.
However, Cifarelli, which is analogous to the claimed invention because it is directed toward identifying fraud using a machine learning model, discloses:
theft risks (Figure 2; paragraphs 0080-0081: Here, fraud is a theft risk)
creating, from the historical event information, positive and negative cases for training the selected time classifier model (paragraphs 0080-0081: Here, a fraud score is calculated from metrics including intermediate scores based upon past metrics derived from histograms (Figure 2, items 220, 221, 230, 231, 232, and 240))
wherein a user remediation occurrence of a fraud alert indicative of identity theft is noted such that the fraud alert is flagged as a positive case for a user, and wherein no user remediation occurrence of the fraud alert indicative of identity theft is noted such that the fraud alert is flagged as a negative case for the user (Figure 2; paragraphs 0071-0089: Here, a fraud alert is identified when the fraud score (S) (Figure 2, item 240) is above a threshold value (Figure 2, items 250 and 251)). Conversely, when the score (S) is not above a threshold value, no alert is generated. Each of the positive cases, a fraud alert is triggered, and the negative cases, no fraud alert is triggered, are provided as input to the machine learning model to train the machine learning model (Figure 2, item 263))
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Cifarelli with Domenikos-Lightbody-Cox-Meyer, with a reasonable expectation of success, as it would have allowed for detecting fraud when a threshold value is reached and further training the model to improve the quality of fraud alerts in real time (Cifarelli: Figure 2; paragraph 0007).
As per dependent claim 2, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Domenikos discloses wherein performing the time series calibration comprises performing one or more of alert compounding, N-day window alert capture, or tri-label encoding (paragraph 0104).
As per dependent claim 3, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 2, and the same rejection is incorporated herein. Domenikos discloses wherein performing the alert compounding comprises considering an alert type a number of times that it appears within a capture window (paragraph 0077).
As per dependent claim 4, Domenikos, Lightbody, Cox, Meyer, and Cifarelli discloses wherein performing the N-day window alert capture comprises considering historical event information from N-days prior to a current data up (paragraph 0077).
As per dependent claim 5, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 2, and the same rejection is incorporated herein. Domenikos discloses wherein the historical event information comprises identity threat alerts (paragraph 0104).
As per dependent claim 6, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 5, and the same rejection is incorporated herein. Domenikos discloses wherein performing the tri-label encoding comprises labeling the historical event information based on user input indicating that an identity thread alert correctly identified a threat or incorrectly identified a threat or labeling the historical event information to indicate that user input was not received for the corresponding alert (paragraph 0156).
As per dependent claim 7, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Domenikos discloses wherein classifying the historical event information comprises graphing alert data and fraud event data against time and deriving, based on the graph, identity score data (paragraphs 0156 and 0158).
As per dependent claim 8, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 7, and the same rejection is incorporated herein. Domenikos discloses wherein displaying the identity interface comprises displaying the graph (paragraphs 0156 and 0158).
As per dependent claim 9, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Domenikos discloses wherein displaying the identity interface comprises displaying a color coded scale and a user’s position on the color coded scale, wherein the user’s position on the color coded scale indicates an identity status for the user (Figure 30; paragraphs 0173-0174).
As per dependent claim 10, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Domenikos discloses wherein the machine learning model is configured to provide real time identify scoring information (paragraphs 0156 and 0158).
As per dependent claim 11, Domenikos, Lightbody, Cox, Meyer, and Cifarelli disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Domenikos discloses wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to train, using neural network regression, a predictive identity scoring model configured to predict identity threat events (Figure 19; paragraph 0270).
With respect to claims 12-19, the applicant discloses the limitations substantially similar to those in claims 1-8, respectively. Claims 12-19 are similarly rejected.
With respect to claim 20, the applicant discloses the limitations substantially similar to those in claim 1. Claim 20 is similarly rejected.
Response to Arguments
Applicant’s arguments with respect to the rejection of claims under 35 USC 101 have been fully considered and are persuasive. The rejection has been withdrawn.
Applicant’s arguments with respect to the rejection of claims 1-20 under 35 USC 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Domenikos, Lightbody, Cox, Meyer, and Cifarelli.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Roth et al. (US 12573029: Discloses generating a training dataset including ground truth data (column 93, lines 51-65)
Kim et al. (US 12482189): Discloses generating a ground truth dataset for training a model (column 89, lines 47-61)
Kadkhodaie Elyaderani et al. (US 2024/0127078): Discloses creating a ground truth dataset by labeling a set of training samples (Figure 8; paragraph 0088)
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 KYLE R STORK whose telephone number is (571)272-4130. The examiner can normally be reached 8am - 2pm; 4pm - 6pm.
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, Omar Fernandez Rivas can be reached at 571/272-2589. 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.
/KYLE R STORK/Primary Examiner, Art Unit 2128
