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 the Claims
Claims 1, 3-10 and 12-19 are all the claims pending in the application.
Claims 2 and 11 are cancelled.
Claims 1, 10, and 19 are amended.
Claims 1, 3-10 and 12-19 are rejected.
The following is a Final Office Action in response to amendments and remarks filed July 21, 2026.
Response to Arguments
Regarding the non-statutory double patenting rejections, the rejections are withdrawn in light of the filed terminal disclaimer.
Regarding the 101 rejections, the rejections are withdrawn because Examiner finds the claims reflect a practical application. That is, under Step 2A Prong 1, Examiner finds the claims recite an abstract idea (i.e., providing career counseling or advice). However under Step 2A Prong 2, Examiner finds the additional elements of using a clustering machine learning model before applying the linear regression based machine learning model, as claimed, reflects an improvement because clustering before applying linear regression as claimed improves accuracy of predictions in data mining1 and reduces the resources needed for training and using the models (i.e., the claims reflect an improvement in machine learning techniques). Accordingly, the 101 rejections are withdrawn.
Regarding the 103 rejections, the rejections are withdrawn at least because the cited references do not teach a link between clustering and linear regression as claimed. Please see below for the new rejections of the claims as amended.
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.
Claim(s) 1, 5, 6, 8-10, 14, 15, and 17-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Xie et al, US Pub. No. 2019/0303798, herein referred to as “Xie”; in view of Fang, US Pub. No. 2017/0286914, herein referred to as “Fang”; further in view of Ari, Bertan, and H. Altay Güvenir "Clustered linear regression." Knowledge-Based Systems 15.3 (2002): 169-175, herein referred to as “Ari”; further in view of Schreiber et al, US Pub. No. 2017/0236095, herein referred to as “Schreiber”.
Regarding claim 1, Xie teaches:
a processor; a non-transient computer-readable medium comprising instructions that when executed by the processor cause the processor to (processor, memory and instructions, e.g., ¶¶ [0068]-[0069], [0073] and Figs. 8 and 9):
receive profile attributes of a plurality of users comprising user data (obtains information from profile database, e.g., ¶¶[0046], [0051] and Figs. 3 and 4; see also ¶[0038] noting profile data includes career related information),
the profile attributes defining a historical progression of actions by each user over a past time period to reach a target state (profile data includes job title, seniority, major, job start date, job send date, years after graduation, etc., ¶[0018], and educational background (e.g., schools, majors, matriculation and/or graduation dates, etc.), employment history, skills, professional organizations, ¶[0038]; see also ¶[0020] discussing identifying career paths taken by users in the training);
cluster, by applying a clustering machine learning model to the profile attributes of the plurality of users to create grouped clusters of users having similar profile attributes within each cluster (clustering algorithm is applied to identify general career paths, ¶¶ [0023], [0047], [0065] and Figs. 3 and 6),
provide, subsequent to the clustering, at least one of the grouped clusters of users to a supervised path determination learning model (clustered notes are passed to career path model component, ¶[0048] and Fig. 3, and career path model component is a supervised machine learning algorithm, ¶¶[0051]-[0052] and Fig. 4)
to apply the profile attributes of a first user from the at least one of the grouped clusters of users associated with a desired target state thereto to estimate a function defining a progression pattern for the historical progression over the past time period to reach the target state (models careers paths, ¶¶[0052]-[0053]; see also ¶[0048] noting model predicts likelihood of candidates proceeding to a career path and ¶[0028] noting recommendations are based on users’ profile and mentors’ profile);
determine for a second user, from a same grouped cluster as the first user, a recommendation for reaching the target state of the first user based on the estimated function and the recommendation comprising a series of actions to be performed to progress to the target state of the first user (uses model to recommend one or more activities designed to enhance the candidate, ¶¶[0048]; see also ¶[0027] noting activities include increasing number of publications in the users profile and adding skills to the profile; and ¶[0059] showing an example of determining how to progress from one position to another);
and trigger a display of the recommendation on a graphical user interface of a client device (recommendations are presented to the user, e.g., ¶¶ [0049], [0055]; see also e.g., ¶ [0073] discussing user devices)
and in response to a selection of the recommendation, provide the input to the clustering machine learning model and the path determination learning model for learning from and adjusting the clustering machine learning model comprising clustering and the supervised path determination learning model in response to the selection for updating subsequent recommendations (alters candidate profile based on activity and resubmits profile to model to revise prediction, ¶¶[0066]-[0067] and Fig. 6).
However Xie does not teach but Fang does teach:
each user represented as a node on an output cluster (includes each member profile is a node, ¶¶[0068]-[0069]; see also, ¶[0036] discussing clustering).
Further, it would have been obvious before the effective filing date of the claimed invention, to combine the career path recommendations of Xie with the social graphs of Fang because known work in one field of endeavor may prompt variations of it for use in the same field based on design incentives, see MPEP 2143.I.F. That is, one of ordinary skill would have recognized modeling career paths would likely be improved by analyzing the connections of the user in addition to their own skills and accordingly would have modified Xie to include the social graphs of Fang.
However the combination of Xie and Fang does not teach but Ari does teach:
provide, subsequent to the clustering, at least one of the grouped clusters a supervised model comprising linear regression (clusters data before applying linear regression, pgs. 171-173).
Further, it would have been obvious before the effective filing date of the claimed invention, to combine the career path recommendations with social graphs of Xie and Fang with the clustered linear regression of Ari because Ari explicitly teaches clustered linear regression improves accuracy of predictions in data, pgs. 169, 175; see also MPEP 2143.I.G.
However the combination of Xie, Fang and Ari does not teach but Schreiber does teach:
a supervised path determination learning model comprising linear regression to apply the profile attributes of a first user (linear regression is performed to choose assess transitions and choose a career path from a source profession to a target profession, ¶¶[0099]-[0100] and Fig. 6).
Further, it would have been obvious before the effective filing date of the claimed invention, to combine the career path recommendations with social graphs and clustered linear regression of Xie, Fang and Ari with the linear regression to determine a career of Schreiber because known work in one field of endeavor may prompt variations of it for use in the same field based on design incentives, see MPEP 2143.I.F. That is, Xie teaches performing the analysis based on logistic regression, ¶[0052]. One of ordinary skill would have modified the teachings of Xie to use linear regression, e.g., as taught by Schreiber, for situations where linear regression is more appropriate (e.g., when trying to predict of continuously distributed outcomes, as opposed to binary outcomes).
Regarding claim 5, the combination of Xie, Fang, Ari and Schreiber teaches all the limitations of claim 1 and Xie further teaches:
update an existing state of the second user to the target state of the first user (alters candidate profile based on activity and resubmits profile to mode to revise prediction, ¶¶ [0066]-[0067] and Fig. 6; e.g., recommends ¶[0027]).
Regarding claim 6, the combination of Xie, Fang, Ari and Schreiber teaches all the limitations of claim 1 and Xie further teaches:
wherein the profile attributes for the users provide metadata characterizing a career profile for each user and further comprises: career related progression patterns for each of the users to move from a first profile state to another along with associated timing information (profile data includes job title, seniority, major, job start date, job send date, years after graduation, etc., ¶ [0018], and graduations dates and employment history, ¶[0038]; see also ¶¶[0019]-[0020] discussing labeling and identifying the career paths in the training data);
training status of each of the users obtained over the past time period (profile data includes educational background, ¶[0038]);
and performance metrics of each of the users within an entity and provided on the graphical user interface and certifications taken by each of the users (analyzed data includes activities weighted based on career path, e.g., publications are weighted more heavily for a professor than a project manager, ¶[0024]).
Regarding claim 8, the combination of Xie, Fang, Ari and Schreiber teaches all the limitations of claim 1 and Xie further teaches:
wherein the recommendations include attributes from the first user's historical progression indicative of a career journey from a prior state to the target state for the first user (uses career paths taken by users in the training data, ¶[0020]),
the attributes including an indication of actions performed comprising: certifications completed, training status changes, and performance metrics (profile data includes job title, seniority, major, job start date, job send date, years after graduation, etc., ¶[0018], and graduations dates and employment history, ¶[0038]).
Regarding claim 9, the combination of Xie, Fang, Ari and Schreiber teaches all the limitations of claim 1 and Xie further teaches:
generate a reasoning for the recommendation and corresponding online actions (determines recommendation by optimizing likelihood of a user corresponding to the candidate user profile progressing down a career path corresponding to the given career path model, ¶[0066]),
the reasoning including features of the historical progression of the first user and an indication of a degree of similarity between the first user and the second user (determines recommendations based on similarities of career paths, ¶[0023]; see also ¶¶[0062]-[0063] discussing identifying a similar cohort).
Regarding claims 10, 14, 15, and 17-19, claims 10, 14, 15, and 17-19 recite similar limitations as claims 1, 5, 6, 8, and 9 and accordingly are rejected for similar reasons as claims 1, 5, 6, 8, and 9.
Claim(s) 3, 4, 7, 12, 13, and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Xie, Fang, Ari and Schreiber, further in view of Bolte, et al, US Pub. No. 2018/0039946, herein referred to as “Bolte”.
Regarding claim 3, the combination of Xie, Fang, Ari and Schreiber teaches all the limitations of claim 1 and does not teach but Bolte does teach:
access a database of available profile states representing potential target states and retrieve associated description metadata (access job sites to obtain job profiles, ¶¶[0065]-[0066]);
and perform natural language processing (NLP) on the description metadata, and the profile attributes of each of the first user and a second user, clustered in a same cluster for being similar to the first user from the at least one of the grouped clusters of users, to determine respective textual context of each (uses natural language processing to extract information and key words from a candidate's resume and from a job description, ¶¶[0112], [0116]).
Further, it would have been obvious before the effective filing date of the claimed invention, to combine the career path recommendations with social graphs and clustered linear regression of Xie, Fang, Ari and Schrieber with the natural language processing based recommendations of Bolte because known work in one field of endeavor may prompt variations of it for use in the same field based on design incentives, see MPEP 2143.I.F. That is, one of ordinary skill would have recognized data about people’s career histories and job descriptions would likely include unstructured text data and accordingly would have used natural language processing to analyze the data when suggesting career paths, e.g., as taught by Bolte.
Regarding claim 4, the combination of Xie, Fang, Ari, Schreiber and Bolte teaches all the limitations of claim 3 and Fang further teaches:
generate the recommendation further based on a determined textual context of the description metadata of a particular role having more than a predefined degree of match with the profile attributes between the first and the second user (determines a degree a resumes matches evaluation data constituting search terms to a job pipeline, ¶[0029]; see also ¶[0025] noting “job pipeline” is a relatively fine grained descriptor of a job classification).
Further, it would have been obvious before the effective filing date of the claimed invention, to combine the career path recommendations of Xie with the matching job descriptions and resumes Fang because known work in one field of endeavor may prompt variations of it for use in the same field based on design incentives, see MPEP 2143.I.F. That is, one of ordinary skill would have recognized modeling career paths would likely be improved by directly comparing job and user resumes, e.g., as taught by Fang and accordingly would have modified Xie to do so.
Regarding claim 7, the combination of Xie, Fang, Ari and Schreiber teaches all the limitations of claim 1 and Fang further teaches:
wherein the recommendation is further generated based on a similarity score determined to the profile attributes of the second user and a description metadata of the target state (determines a degree a resumes matches evaluation data constituting search terms to a job pipeline, ¶[0029]; see also ¶[0025] noting “job pipeline” is a relatively fine grained descriptor of a job classification).
Further, it would have been obvious before the effective filing date of the claimed invention, to combine the career path recommendations of Xie with the matching job descriptions and resumes Fang because known work in one field of endeavor may prompt variations of it for use in the same field based on design incentives, see MPEP 2143.I.F. That is, one of ordinary skill would have recognized modeling career paths would likely be improved by directly comparing job and user resumes, e.g., as taught by Fang and accordingly would have modified Xie to do so.
However the combination of Xie, Fang, Ari and Schreiber does not teach but Bolte does teach:
by applying natural language processing (uses natural language processing to extract information and key words from a candidate's resume and from a job description, ¶¶[0112], [0116]).
Further, it would have been obvious before the effective filing date of the claimed invention, to combine the career path recommendations with social graphs and clustered linear regression of Xie, Fang, Ari and Schrieber with the natural language processing based recommendations of Bolte because known work in one field of endeavor may prompt variations of it for use in the same field based on design incentives, see MPEP 2143.I.F. That is, one of ordinary skill would have recognized data about people’s career histories and job descriptions would likely include unstructured text data and accordingly would have used natural language processing to analyze the data when suggesting career paths, e.g., as taught by Bolte.
Regarding claims 12, 13, and 16, claims 12, 13, and 16 recite similar limitations as claims 3, 4, and 7 and accordingly are rejected for similar reasons as claims 3, 4, and 7.
Conclusion
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 BRENDAN S O'SHEA whose telephone number is (571)270-1064. The examiner can normally be reached Monday to Friday 11-7.
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.
/BRENDAN S O'SHEA/Examiner, Art Unit 3626
1 See new reference Ari, pgs. 169, 175
Prosecution Insights