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 the Claims
Claims 1-20 were previously pending and subject to a final office action mailed 02/26/2026. Claims 1 and 11 were amended; no claim was cancelled, or added in a reply filed 05/21/2026. Therefore claims 1-20 are currently pending and subject to the non-final office action below.
Response to Arguments
Applicant argues “In Desjardins, the ARP reviewed a Board panel decision that entered a new ground of rejection of claims 1-6 and 8-20 under § 101. The ARP vacated the Board's new ground of rejection. The ARP stated that the panel "essentially equated any machine learning with an unpatentable 'algorithm' and the remaining additional elements as 'generic computer components,' without adequate explanation." Desjardins ARP Decision, p. 9. The ARP further stated that "Categorically excluding Al innovations from patent protection in the United States jeopardizes America's leadership in this critical emerging technology" and that "Examiners and panels should not evaluate claims at such a high level of generality." Desjardins ARP Decision, pp. 8-9.
The Office's maintained rejection treats machine-learning elements and score-related elements at the same high level of generality that Desjardins cautions against. The USPTO Memo provides the appropriate analysis. The USPTO Memo explains that "The claim limitation 'training the neural network in a first stage using the first training set' of example 39 does not recite a judicial exception. Even though 'training the neural network' involves a broad array of techniques and/or activities that may involve or rely upon mathematical concepts, the limitation does not set forth or describe any mathematical relationships, calculations, formulas, or equations using words or mathematical symbols. Contrast this with the limitation 'training, by the computer, the ANN based on the input data and a selected training algorithm to generate a trained ANN, wherein the selected training algorithm includes a backpropagation algorithm and a gradient descent algorithm' of claim 2 of example 47. This limitation requires specific mathematical calculations by referring to the mathematical calculations by name, i.e., a backpropagation algorithm and a gradient descent algorithm, and therefore recites a judicial exception, namely an abstract idea." USPTO Memo, p. 2.
Amended claim 1 recites "generating a first temporary ID for the at least a portion of the aggregated job candidate data," "identifying a first static ID associated with the prospective job candidate," "storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID," "training, based on the aggregated job candidate data for the first job type and corresponding outcome labels comprising one or more of hiring outcomes and role-transition events, a machine learning model configured to output a job seeker skillset match score as a bounded scalar," and "determining, via the trained machine learning model based on a comparison of current skills possessed by the prospective job candidate and the relevant skills, the job seeker skillset match score." No recitation identifies a backpropagation algorithm, gradient descent, or any other specific mathematical calculation by name.
The Office also has not explained how the recitation "storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID" sets forth a mathematical relationship, calculation, formula, or equation using words or mathematical symbols. The rejection therefore does not establish that the amended claims recite a mathematical concept under Step 2A, Prong One. “(remarks p. 8-9). Examiner respectfully disagrees.
The mathematical concept basis does not rest on the training step, it rests on specific quantitative claim language. Applicant’s Example 39 argument applies that example’s conclusion to the claim as a whole, but Example 39 involved a claim whose only relevant output was an improved neural network for facial detection, there was no recitation of a numeric score, no recitation of a bounded output domain, and no recitation of a comparison producing a quantitative result. The present claim is distinguishable on its face. MPEP 2106.04(a)(2)(I) defines mathematical concepts to include “mathematical relationships, mathematical formulas or equations, mathematical calculations” The mathematical concept basis here does not rest on the training step naming a specific algorithm, it rests on three specific recitations that use quantitative and mathematical language.
First, “a machine learning model configured to output a job seeker skillset match score as a bounded scalar”. The term “bounded scalar” is not a functional description of a business outcome. It is a mathematical characterization of the output’s structure, a single numerical value constrained within a finite numeric domain. Reciting that the output must satisfy a mathematical property is a recitation of a mathematical relationship between the models’ output and a defined numeric range. This is not equivalent to Example 39’s “training the neural network in a first stage using the first training set” which recited nothing about the mathematical character of the output. The claim here goes further by specifying the mathematical nature of what the model must produce.
Second, “determining, via the trained machine learning model based on a comparison of current skills possessed by the prospective job candidate and the relevant skills, the job seeker skillset match score” A comparison of two data sets that produces a numeric score is a mathematical calculation. The claim does not merely say “use the model to produce a recommendation”; it specifies that the score is determined based on a comparison, a mathematical operation taking two inputs and producing a numeric result. MPEP 2106.04(a)(2)(I) expressly covers mathematical calculations, and a comparison producing a numeric score is a calculation regardless of whether the underlying algorithm is named.
Third, the dependent claims recite “ranking the one or more relevant skills based on a quantity of the plurality of job candidates having a respective skill, an amount of experience” Ranking items by numeric quantities is an explicit mathematical operation. The terms “quantity” and “amount” are mathematical terms, and ordering by those values is a mathematical calculation. This is not a generalization imposed from outside the claim, it is what the claim language expressly recites.
The example 39 domain context distinction is not merely assertion, it is grounded in what Example 39 actually analyzed. Applicant argues that the USPTO Memo does not say the Example 39 analysis is domain specific, and that the examiner’s domain context distinction is unsupported. The distinction, however, does not rest on domain alone. It rests on what the claim in Example 39 actually recited versus what the present claim recites. Example 39’s claim did not recite a bounded scalar output, did not recite a comparison producing a numeric score, and did not recite ranking by numeric quantity. The present claim recites all three. The domain point was secondary to the primary distinction which is that the claims are not analogous because they do not recite the same type of limitations. Example 39 cannot be transplanted to the present claim because the present claim recites quantitative output and calculation language that Example 39 did not contain.
The persistent cache limitation does not bear on the mathematical concept basis. The persistent cache limitation is an additional limitation. Examiner does not content that storing an association record in a persistent cache recites a mathematical concept. The mathematical concept basis is directed solely to the bounded scalar output and the skill comparison producing a numeric score. Applicant’s persistent cache argument addresses a position the examiner does not hold.
Applicant invokes Desjardins for the proposition that the Office is treating ML elements at too high a level of generality. Desjarsins cautioned against categorically equating machine learning with an unpatentable algorithm without adequate explanation. The Examiner does not make that categorical equation. The mathematical concept basis is directed to specific claim language (i.e. bounded scalar, “comparison”, producing a score, quantity, amount, ranking) that uses quantitative, mathematical terms. That is a limitation specific analysis, not a categorical treatment of machine learning. Desjardin’s instruction is followed, not violated.
Applicant argues “The Office states that, although certain computing steps cannot be practically performed in the human mind in their entirety, "the substance of what the method accomplishes, identifying candidate skill gaps, comparing those gaps to job requirements, and recommending a more suitable job type, is precisely the kind of evaluation and recommendation that human recruiters and career counselors have long performed mentally and on paper." Office Action, p. 8. Applicant respectfully submits that this approach analyzes the claim at too high a level of generality and does not account for the specific amended recitations. The USPTO Memo expressly cautions against that type of overgeneralization. The USPTO Memo states that "Examiners are reminded not to expand this grouping in a manner that encompasses claim limitations that cannot practically be performed in the human mind." USPTO Memo, p. 2. The USPTO Memo further states that "Claim limitations that encompass AI in a way that cannot be practically performed in the human mind do not fall within this grouping." Id. Here, the amended claims are not limited to a human evaluation of a candidate or a career- counseling recommendation. Amended claim 1 recites "generating a first temporary ID for the at least a portion of the aggregated job candidate data," "identifying a first static ID associated with the prospective job candidate," and "storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID." Those recitations are part of a computer-implemented data-handling sequence that also recites ATS/HRIS data retrieval, model training, and model-based scoring. The Office has not shown how the persistent-cache association between the first temporary ID and the first static ID can practically be performed in the human mind, and it has not shown how a human mind could practically perform the ordered computer-implemented sequence recited in the amended claims. Characterizing the result as candidate evaluation does not satisfy Step 2A, Prong One, because Prong One turns on what the claim recites, not on a generalized description of a business purpose.
That conclusion is consistent with Desjardins. The ARP cautioned that "Examiners and panels should not evaluate claims at such a high level of generality." Desjardins ARP Decision, p. 9. The Office's mental-process reasoning does just that by treating the amended claims as the idea of evaluating a candidate while overlooking the recited persistent-cache association, the recited ATS/HRIS data retrieval, and the recited trained machine learning model.” (remarks p. 9-10).
Examiner respectfully disagrees. MPEP 2106.04(a)(2)(III) provides that mental processes include “concepts performed in the human mind, including observations, evaluations, judgments, and opinions” and that a claim limitation falls within this grouping when it “can practically be performed in the human mind.” The MPEP further provides that the mental process grouping is not limited to steps literally performed in the mind but extends to steps that can practically be performed mentally, including with the aid of pen and paper. The question for each recitation is therefore whether a human being, with the assistance of pen and paper if needed, could practically perform the recited steps.
The identifier management limitations can practically be performed in the human mind with pen and paper. Generating a first temporary ID for the at least a portion of the aggregated job candidate data recites the act of assigning a temporary reference token to a set of retrieved data. A human HR clerk or recruiter has performed this exact operation for decades, assigning a batch number, a file reference code, or a temporary case number to a set of candidate records pulled from a filing system. Writing “TMP 001” at the top of a folder of pulled candidate files is the human analog of this step. It requires no computing infrastructure, no AI, and no technical implementation that is beyond human cognitive and manual capability. It is a classic administrative act performable mentally or with pen and paper.
Identifying a first static ID associated with the prospective job candidate recites the act of looking up or recognizing a persistent identifier, an employee ID, a candidate number, a social security number, or any other stable reference, associated with a specific individual. A human HR clerk performs this step every time they pull a candidate’s permanent file and note their employee or applicant ID number. It is a lookup and recognition task that requires no AI, no model, and no computing capability beyond human memory or a manual filing system. It is performable mentally or with pen and paper.
Storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID recites the act of recording that a specific temporary reference is currently associated with a specific persistent identifier. The human analog is writing “TMP 001 = candidate jane DOE, applicant ID 45” in a logbook or on an index card and placing it in a filing cabinet for later reference. Human administrative workers have maintained exactly this type of cross reference record in paper-based HR systems for as long as organized HR functions have existed. A “persistent cache” in the human analog is a filing cabinet, a logbook, or an index card box, a persistent physical storage medium that maintains the association for later retrieval. None of this requires computing infrastructure or AI.
The USPTO Memo’s instructions that “claim limitation that encompass AI in a way that cannot be practically performed in the human mind do not fall within this grouping” applies specifically to AI driven operations, model training, inference at scale, neural network execution. None of the three identifier management limitations encompass AI. They are administrative data management steps that are plainly performable by a human with pen and paper.
Similarly, the core employment matching steps can practically be performed in the human mind with pen and paper. Beyond the identifier management limitations, the prior Office Action response correctly identifier that the substance of what the claim accomplishes, receiving a candidate request for a job type, assessing the candidate’s skill fitness, and recommending an alternate career path based on missing skills, is precisely what human recruiters and career counselors have long performed mentally and on paper. That observation stands and is not undermined by the amendment.
The following two steps that cannot practically be performed in the human mind are acknowledged but they dot no control the analysis. “Retrieving, via an API, at least a portion of the aggregated job candidate data from an ATS and/or HRIS”: programmatic API based retrieval of enterprise HR database records cannot practically be performed mentally; “training, based on the aggregated job candidate data for the first job type and corresponding outcome labels, a machine learning model”: training a machine learning model on large scale historical hiring outcome data cannot practically be performed in the human mind. However, the presence of two non-mental steps within a claim does not automatically remove the claim from the mental process grouping or establish that the claim as a whole is not directed to a mental process. MPEP 2106.04(a)(2)(III) does not require that every single limitation be performable mentally; it requires evaluation of whether the claim recites a mental process concept. When the claim’s core operative steps, receiving a candidate’s request, assessing skill fitness, identifying missing skills, and recommending an alternate career path, are all performable mentally or with pen and paper, and when even the amended identifier management steps are performable mentally or with pen and paper as demonstrated above, the presence of an API retrieval step and a model training step does not change the overall character of what the claim recites.
Furthermore, the API retrieval steps and model training step are additional elements that are evaluated under prong two and step 2B, not as the core of the abstract idea itself. Their presence in the claim as computer implemented components does not establish that the claim is directed to a technical improvement rather than to the mental process of evaluating and recommending employment paths.
Therefore, the mental process characterization is not a reduction of the claim to a vague business purpose. It is grounded in a specific limitation by limitation analysis, as set forth above, demonstrating that each of the claim’s core operative steps, including the amended identifier management limitations, can practically be performed by a human with pen and paper. That is the opposite of overgeneralization. It is a more granular, limitation specific analysis than Applicant’s own argument, which broadly asserts that the amended claim “cannot be performed in the human mind” without engaging with the individual recitations.
Applicant argues “The Office alleges that claim 1/11, under its broadest reasonable interpretation, covers a method of organizing human activity. Office Action, pp. 15-16. The Office further states that the method allows for "commercial or legal interactions" including "business relations" and for "managing personal behavior or relationships or interactions between people." Office Action, p. 16. Applicant respectfully disagrees.
The organizing-human-activity grouping is not a catchall for every claim that has some relationship to people or employment. The relevant inquiry is whether the claim recites one of the recognized categories of certain methods of organizing human activity. The amended claims do not recite a contractual relationship, a legal obligation, an advertising activity, a sales activity, a rule for human conduct, social activity, teaching activity, voting activity, or game-play activity. Nor does the fact that the claimed system operates in an employment-related environment transform the specific recitations into a method of organizing human activity. Amended claim 1 recites "generating a first temporary ID for the at least a portion of the aggregated job candidate data," "identifying a first static ID associated with the prospective job candidate," and "storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID." These recitations address how the computer system handles and associates identifiers for retrieved data, not a commercial agreement, legal relationship, interpersonal relationship, or rule for human behavior.
The Office's reasoning also conflicts with the USPTO Memo's instruction to distinguish claims that recite an exception from claims that merely involve one. The USPTO Memo states that "Examiners should be careful to distinguish claims that recite an exception (which require further eligibility analysis) from claims that merely involve an exception (which are eligible and do not require further eligibility analysis)." USPTO Memo, p. 2. Even if the claims involve an employment context, the amended recitations do not recite commercial or legal interactions or managing personal behavior or relationships.
Accordingly, the Office has not established that the amended claims recite certain methods of organizing human activity under Step 2A, Prong One. The rejection should not be maintained on that abstract-idea grouping.” (remarks p. 10-11).
Examiner respectfully disagrees. Examiner respectfully notes that the sub-categories are not exhaustive lists. MPEP 2106.04(a)(2)(III) identifies “commercial or legal interactions” and “business relations” as sub-categories of certain methods of organizing human activity. These are not exhaustive enumerations. The MPEP identifies contracts, legal obligations, advertising, sales, and similar activities as examples within those sub-categories, not as the complete boundaries of them. The relevant question is whether the claim recites a commercial interactions or a business relation in substance, not whether it uses the vocabulary of one of the named examples.
The claim as a whole recites a commercial employment matching workflow. Reading the independent claim as a whole under BRI demonstrates that the claim is not merely a claim that involves employment in some peripheral way. It is a claim that recites the complete operational sequence of a commercial employment matching and career advisory service, receiving, requests from job seekers, assessing their fitness for employment categories, and recommending career paths. Those are quintessential commercial interactions in the employment domain and business relations between the platform, candidates, and the labor market.
The USPTO’s Memo’s “recite vs. involve” distinction does not help Applicant here. The Memo’s instruction to distinguish claims that “recite” an exception from those that “merely involve” one is directed at preventing examiners from identifying abstract ideas based on tangential or peripheral relationships to known abstract concepts, for example, finding a claim abstract because it processes data, or because it involves a computer. That instruction does not mean that a claim whose core operative steps describe a commercial workflow fails to recite an organizing human activity exception merely because those steps are implemented on a computer. The present claim does not merely involve employment in a peripheral way; its operative steps, receiving candidate requests, assessing skill fitness, and recommending career paths, are the commercial activity. The computer implementation is the means, not the substance.
The identifier management limtiations do not change the Prong One character of the claim. Applicant argues that the temporary ID, static ID, and persisten cache limitations describe how the computer handles data identifiers, not a commercial agreement or relationship. This is accurate as a description of those limitaitons in isolation. But Prong One requires evaluation of the claim as a whole, not each limitation in isolation. (see MPEP 2106.04(a)). The identifier-management limitations are embedded within and in service of the employment matching workflow. They do not redirect the claim toward a technical computing problem or a technical computing solution. Tey manage data identity within a pipeline whose purpose and operative output is commercial employment matching. The overall character of the claim, evaluated as a whole, remains a method of organizing commercial employment interactions.
Applicant argues “Even if the claims were found to recite a judicial exception, the analysis moves to Prong Two of Step 2A. Under Prong Two, the Examiner evaluates whether the claim as a whole integrates the recited judicial exception into a practical application of the exception. See MPEP § 2106.04(d). The USPTO Memo states that "[t]he analysis in Step 2A Prong Two considers the claim as a whole" and that "[t]he way in which the additional elements use or interact with the exception may integrate the judicial exception into a practical application." USPTO Memo, p. 3. The USPTO Memo further instructs that "the analysis should take into consideration all the claim limitations and how these limitations interact and impact each other when evaluating whether the exception is integrated into a practical application." Id.
Applicant respectfully submits that the amended claims integrate any alleged judicial exception into a practical application under the USPTO Memo and the Desjardins ARP Decision. The Office's analysis does not account for the amended claims as a whole or for the way the recited persistent-cache association interacts with the ATS/HRIS retrieval, model training, model-based scoring, and recommendation recitations.
The Office asserts that "Each of the additional limitations is recited at a high level of
generality and amounts to no more than mere instructions to apply the exception using a generic computer component," and that "Accordingly, these additional elements, alone or in combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea." Office Action, pp. 16-17. The amended claims address that criticism by reciting a particular persistent-cache association in the system's data flow.
The USPTO Memo instructs that an important consideration is "the extent to which the claim covers a particular solution to a problem or a particular way to achieve a desired outcome, as opposed to merely claiming the idea of a solution or outcome." USPTO Memo, p. 3. The USPTO Memo also instructs examiners to consider whether the claim invokes computers merely as a tool or instead purports to improve computer capabilities or another technology. Id. Desjardins likewise recognizes that claim recitations may reflect an "improvement to how the machine learning model itself operates." Desjardins ARP Decision, p. 8.
i. The claims provide an improvement to technology
The USPTO Memo explains that, in Step 2A, Prong Two, the analysis "considers the
claim as a whole" and that claims may be eligible where "a claim reflects an improvement to the functioning of a computer or to another technology or technical field." USPTO Memo, pp. 3-4. Desjardins likewise states that claims "directed to an improvement in the functioning of a computer, or an improvement to other technology or technical field are patent eligible." Desjardins, p. 7. Desjardins further states that an assertion in the specification "alone is insufficient to support a patent eligibility determination, absent a subsequent determination that the claim itself reflects the disclosed improvement." Id., pp. 7-8.
a. Identification of the Technology Field
The technology field is computer-implemented identity management and data integration in ATS/HRIS-supported career management co-op platform systems. The specification describes a CMCP data management and storage system that "may integrate with existing ATS and HRIS systems through API connections," and that the digital resume or wallet may serve as a central repository for career-related data. Specification, 32. The specification also describes computing devices for implementing ATS, employer system, and CMCP functionality. Specification, 87- 88.
b. Particular Improvement
The particular improvement is reflected in the recitations "generating a first temporary ID for the at least a portion of the aggregated job candidate data," "identifying a first static ID associated with the prospective job candidate," and "storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID." These recitations provide a particular way for the computer system to maintain identity continuity between retrieved aggregated job candidate data and the prospective job candidate before the trained model determines the job seeker skillset match score.
This improvement is not directed to the abstract idea of recommending a job. It is
directed to how the computer-implemented CMCP system preserves an association between a temporary identifier for retrieved data and a static identifier associated with the prospective job candidate within a persistent cache. In that respect, the amended claims now recite technical particulars that the Office previously found missing when it stated that the claims did not describe a technical mechanism by which API retrieval from an ATS/HRIS operates differently from conventional database queries. Office Action, p. 11.
c. Clear Difference from the Conventional Approach and How the Claims Deliver the Improvement
To appreciate the clear difference between the amended claims and conventional approaches, the relevant shortcomings should be considered from the perspective of the technology field. Conventional resumes and CVs "often fail to capture the dynamic nature of an individual's skills and experience," creating a need for a solution that can "dynamically manage and display a job seeker's career data." Specification, 3. The specification describes a CMCP system that aggregates data from client ATSs, job boards, jobseekers directly, HRIS, or any combination thereof, and manages a digital resume or wallet that may aggregate an individual's work history, skills, qualifications, and certifications. Specification, 4-5.
Those shortcomings are not merely business concerns. They arise from the way career- related data is collected, stored, validated, updated, and reused across different systems. The specification describes a digital resume or wallet that serves as a central repository for career- related data, including work history, skills, certifications, and soft skills, and further describes API-based integration with existing ATS and HRIS systems. Specification, 32. The specification also describes candidate profile data that includes a digital resume or wallet, machine-learning inference for analyzing aggregated candidate profile data, and a recommendation system that provides recommendations based on calculated candidate and job scores. Specification, 35-37.
The shortcomings of conventional career-data and matching systems
Conventional approaches that rely on resumes, CVs, or isolated applicant records are poorly suited to dynamic career data because they do not provide a particular mechanism for maintaining a current association between externally retrieved data and a persistent candidate identity as data moves into downstream scoring or recommendation processes. The specification supports that the CMCP environment involves digital resume or wallet data, verified work- history validation, static and non-static credentials, and application history with status updates and feedback mechanisms. Specification, 38-41. The specification further describes that the digital resume or wallet may be updated in real time as job seekers acquire new skills, certificates, or work experience, and that the AI module may keep learning with each candidate and the data gathered throughout the process. Specification, 45-46.
2. How the claims deliver the improvement
Amended claim 1 recites "retrieving, via an application programming interface (API), at least a portion of the aggregated job candidate data from an applicant tracking system (ATS) and/or a human resource information system (HRIS)," "generating a first temporary ID for the at least a portion of the aggregated job candidate data," "identifying a first static ID associated with the prospective job candidate," and "storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID." The claim then recites training the machine learning model and determining the job seeker skillset match score via the trained machine learning model. The quoted recitations provide a specific data-handling pathway that addresses the conventional shortcoming described above by maintaining identity continuity between retrieved aggregated job candidate data and the prospective job candidate before the model-training and model-based scoring operations occur. The specification supports the surrounding data flow by describing retrieval of aggregated job candidate data from employer systems, ATS, job board systems, and employer systems, as well as determining a job seeker skillset match score via machine learning inference. Specification, 97-98.
d. All Embodiments Within the Claim Scope Result in the Improvement
Every embodiment within the scope of amended claim 1 includes the recitations quoted above. Because amended claim 1 recites the persistent-cache association between the first temporary ID and the first static ID as part of the data flow, the asserted improvement is intrinsic to the claim structure and not dependent on any unclaimed variable. The corresponding system recitations in amended claim 11 provide the same identity-association mechanism.
Practical Application and Integration
The improvement described above is integrated into a practical application with a direct nexus between the persistent-cache association and the downstream operation of the CMCP system. The persistent-cache association is not appended after the alleged abstract idea. It is recited before the model training and model-based scoring recitations and forms part of the data- handling pathway used by the system.
The USPTO Memo states that, "[w]hile an additional limitation (or combination) that merely applies the judicial exception on a generic computer may not render a claim eligible on its own, an additional limitation (or combination) that meaningfully limits the judicial exception can render it eligible." USPTO Memo, p. 3. The persistent-cache association meaningfully limits any alleged exception because it confines the claim to a particular data-integration mechanism used within the CMCP system, rather than to the generalized concept of evaluating a candidate or recommending a job.
The claims therefore do not merely state a desired result. They recite a particular data- handling mechanism and integrate that mechanism with ATS/HRIS retrieval, model training, model-based scoring, and a system-provided recommendation. Accordingly, Applicant respectfully submits that, even if Step 2A, Prong One were satisfied, the amended claims integrate any alleged judicial exception into a practical application under Step 2A, Prong Two.” (remarks p. 11-15).
Examiner respectfully disagrees.
Applicant argues that the prior Office action found the calims deficient because they lacked a technical mechanism describing how API retrieval from an ATS/HRIS operates differently from conventional database queries, and that the amendment responds directly to that deficiency by adding the identifier management limitations. This argument requires direct answer.
The prior Office Action’s observation about the absence of a technical mechanism for ATS/HRIS retrieval was directed at the claim’s failure to recite any technically specific or non-conventional implementation of the retrieval step itself. the amendment addresses the retrieval step not by specifying how retrieval operates differently, but by adding downstream identifier management steps that operate on the retrieved data. The amendment does not cure the previously identified deficiency in the retrieval step; it adds a different step. The retrieval step remains recited at the same level of generality, “retrieving, via an application programming interface (API), at least a portion of the aggregated job candidate data from an application tracking system (ATS) and/or a human resource information system (HRIS), without any specification of a non-conventional retrieval mechanism, protocol or architecture.
Furthermore, even if the identifier management limitations are treated as the responsive addition, the question for Prong two is not whether the amendment responds to a prior office observation but whether the amended claim as a hwole integrates any alleged judicial exception into a practical application. The amendment’s responsiveness to a prior observation does not automatically satisfy Prong Two. The analysis must still evaluate whether the added limitations reflect a technical improvement, a particular solution to a technical problem, or a non-conventional application of the exception. For the reasons set forth below, they do not.
The persistent cache limitation does not reflect an improvement to computer functionality. MPEP 2106.05(a) requires that a technical improvement be reflected in the claim language itself, the claim must recite the details of how the computer is improved, the extent to which the computer aids the method, or the significance of the computer’s role is achieving the improvement. The persistent cache limitation recites: a data structure, a record; the content of that record, a current association between a first temporary ID and a first static ID; and the storage location, a persistent cache. The limitation does not recite how the persistent cache is technically implemented or why it is superior to alternative storage structure; how the temporary ID is generated in a technically non-routine manner; how the static ID is identified in a technically unconventional way; how the association record is maintained or updated in a way that improves data integrity, system performance, or computing efficiency; or what technical problem arising from the computer system’s own architecture is solved by this specific storage mechanism.
The claim recites what is stored, not how the storage improves the computer. That distinction is controlling under MPEP 2106.05(a). A claim that recites storing a particular data structure in a conventional storage location describes a desired functional result, the association is stored without specifying any technically unconventional means of achieving it.
The technology field identification does not change the analysis. Applicant identifies the technology field as “computer implemented identity management and data integration in ATS/HRIS supported career management co-op platform systems” The Examiner acknowledges that the claims operate in a computing environment that involves identity management and data integration However, identifying the technology field broadly does not establish a technical improvement within that field. the Prong Two question is not whether the claim operates in a technical field but whether the claim reflects an improvement to that technical field as recited in the claim language. (see MPEP 2106.05(a)). The claim’s recitation of a persistent cache storing an identifier association does not reflect an improvement to identity management technology, data integration technology, or ATS/HRIS integration technology. It describes a conventional data management operation pairing a temporary token with a persistent identifier and storing the pair, that is standard practice in systems that retrieve and process data from external sources on behalf of individual users.
The USPTO Memo instructs that Prong Two considers “the extent to which the claim covers a particular solution to a problem or a particular way to achieve a desired outcome, as opposed to merely claiming the idea of a solution or outcome” Applicant argues that the persistent cache association, combined with ATS/HRIS retrieval, model training, model based scoring, and a maintaining identity continuity across a multi step data pipeline.
The particular solution consideration in the Memo is directed at distinguishing claims that are confined to a specific technical implementation of the exception from claims that broadly cover any implementation of the exception. A claim is a “particular solution” in the relevant sense when its recitations limit it to a specific technical approach that is distinguishable from other ways of achieving the same outcome. The question is therefore: does the persistent cache association, as recited, limit the claim to a specific technical approach that is distinguishable from other ways of maintaining identity continuity in a data pipeline?
And it does not. The recitation “storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID” does not specify how the temporary ID is generated, what makes it temporary, how the static ID is identified, what the persistent cache’s technical characteristics are, or how the association record is structured or maintained. Any system that generates any temporary token for any retrieved data, pairs it with any persistent user identifier, and stores any record of that pairing in any persistent storage structure falls within the scope of this limitation. That is not a particular solution to a technical problem; it is a broad functional description of a category of conventional data management operations. The claim covers the idea of maintaining an identifier association in a cache, not a specific technical implementation of that idea. The Memo’s “particular solution” consideration therefore does not support eligibility here.
Applicant argues that because every embodiment within the scope of the amended claim includes the persistent cache association, the improvement is intrinsic to the claim structure and not dependent on any unclaimed variable. The fact that a limitation appears in every embodiment of a claim does not establish that the limitation integrates the judicial exception into a practical application. The prong two analysis asks whether the limitation meaningfully limits the exception, whether it reflects a non-abstract application of the caption across all claim scope. A limitation that is present in every embodiment but that describes a conventional data management operation in every embodiment does not become a meaningful limitation on judicial exception merely because it is universally present. The “all embodiments” argument conflates the presence of a limitation with its substantive contribution to the Prong Two analysis.
Applicant argues that McRo, DDR Holdings, BASCOM, Amdocs and EX parte Smith support eligibility for the amended claim, not merely the pre-amendment claim, and that the amendment changes the analysis under each. Each is addressed in light of the amendment.
In McRo, the Federal Circuit found eligibility because the claims recited specific rules, defined parameters and transition weightings, that produced a technical result (3D lip synchronization) that was previously unachievable and that the rules defined in a technically specific, previously unavailable way. The amendment does not add anything analogous to the specific rules in McRo. the persistent cache association does not define specific technical parameters, procedures, or rules for how the employment matching pipeline operates. It adds a data management step that is described at the same functional level of generality as the rest of the claim. The amendment does not bring the claim within McRo’s rationale because it does not add the kind of technically specific procedural rules that McRo found dispositive.
In DDR Holdings, the Federal Circuit found eligibility because the claims addressed a problem unique to the Internet and solved it through a specific architectural modification to how web servers generated and served composite pages. The persistent cache association does not address a problem unique to the computer network environment or to the specific technical architecture of ATS/HRIS integration. The problem Applicant identifies, maintaining identity continuity across a data pipeline, is not a problem unique to computing. Data pipelines require identity management in any context, and the solution recited, storing an identifier pairing in a cache, is a conventional solution to a conventional data management challenge.
In BASCOM eligibility rested on a non-conventional arrangement of an internet content filter at a specific network location, the ISP rather than a local or remote location, which addressed a specific technical tradeoff inherent in the network architecture. The amendment does not add a non-conventional arrangement of components. The persistent cache is described at a generic level without specifiying its placement in the system architecture, its relationshnip to other components, or any technical tradeoff it addresses. The arrangement of components in the amended claim (i.e. device, API, ATS/HRIS, persistent cache, ML model, output device) is a conventional client server data processing architecture.
In Amdocs, the Federal Circuit found eligibility based on a distributed processing architecture in which each network device processed its own data near the source to address a network data volume problem specific to the system’s architecture. The amendment does not add a distributed processing architecture or any analogous non-conventional component arrangement. The persistent cache is recited as a storage location for an association record, not as apart of a distributed architecture that addresses a technical data volume or processing challenge.
In Ex parte Smith, the PTAB found eligibility for a trading platform that used specific timers, event driven workflows, and cross system order routing to address latency problems that were technical in nature and specific to the high speed electronic trading environment. The persistent cache association does not address a latency problem, a throughput problem, or any other technical performance problem arising from the computing or network architecture of the CMCP system. It addresses a data management concern, maintaining identity continuity, that is a business requirement of the employment matching workflow, not a technical challenge arising from the system’s computing architecture.
Applicant argues “Under Step 2B, the Examiner considers whether the additional elements recited in the claim represent significantly more than a judicial exception. The Office asserts that the additional elements are "nothing more than mere instructions to apply the exception on a general computer." Office Action, p. 17. Applicant respectfully disagrees.
When viewed as an ordered combination, the amended claims recite more than the alleged abstract idea. The amended claims recite, among other elements, "storing, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID," together with ATS/HRIS data retrieval, training a machine learning model using outcome labels, and determining a job seeker skillset match score via the trained machine learning model. These elements, considered individually and in combination, support a conclusion that the claims recite significantly more than any alleged abstract idea. Accordingly, Applicant respectfully submits that the rejection under 35 U.S.C. § 101 should be withdrawn. “ (remarks p. 15). Examiner respectfully disagrees.
The ordered combination does not supply significantly more. Applicant argues that the ordered combination (ATS/HRIS retrieval feding into identifier management, which feeds into model training, which feeds into model based scoring, which feeds into a device delivered recommendation) constitutes a specific technical architecture that has not been shown to exists as a combined system in the prior art. This argument conflates the 103 analysis with the step 2B analysis. Step 2B does not require that the exact combination of elements appear in a single prior art reference. MPEP 2106.05(d) asks whether the additional elements, individually and in combination, represent well-understood, routine, and conventional activity. The combination of individually conventional steps in their conventional roles, retrieving data via API, managing identifiers in a cache, training a model on labeled data, scoring a candidate, and delivering a recommendation, is itself a conventional sequential data processing pipeline. The fact that each step feeds into the next in logical order does not transform the ordered combination into a n inventive concept. Sequential pipelines of conventional computing steps are the paradigmatic example of what step 2B is designed to screen out (please see MPEP 2106.05(d)).
Therefore, the claims limitation, alone or in combination, do not provide significantly more.
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 claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Claim 1/11 is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim “receiving a request for consideration for a first job type for the job candidate; determining, based on aggregated job candidate data of a plurality of job candidates having the first job type, one or more relevant skills possessed by the plurality job candidates; generating a first temporary ID for the at least a portion of the aggregated job candidate data; identifying a first static ID associated with the prospective job candidate; storing a record indicating a current association between the first temporary ID and the first static ID; retrieving at least a portion of the aggregated job candidate data; determining, via machine learning model, based on a comparison of current skills possessed by the prospective job candidate and the relevant skills, a job seeker skillset match score; and based on the skillset match score satisfying a threshold, providing [prospective job candidate] a recommendation of a second job type relevant to at least one of the one or more relevant skills missing from the current skills of the prospective job candidate.”
The limitations above, as drafted, is a process that, under its broadest reasonable interpretation, covers a method of organizing a human activity and mathematical concepts. That is, the method allows for fundamental economic principles or practices (including hedging, insurance, mitigating risk); commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations) and managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions) and mathematical relationships.
This judicial exception is not integrated into a practical application. In particular, the claim recites a device associated with a prospective job candidate, an application programming interface (SPI), an applicant tracking system (ATS) and/or a human resource information system (HRIS), “training, based on the aggregated job candidate data for the first job type and corresponding outcome labels comprising one or more of hiring outcomes and role- transition events, a machine learning model configured to output a job seeker skillset match score as a bounded scalar “ and persistent cache (claim 1), and a device associated with a prospective job candidate configured to provide a request for consideration for a first job type for the job candidate; and a career management co-op platform (CMCP) system, an application programming interface (SPI), an applicant tracking system (ATS) and/or a human resource information system (HRIS), “train, based on the aggregated job candidate data for the first job type and corresponding outcome labels comprising one or more of hiring outcomes and role- transition events, a machine learning model configured to output a job seeker skillset match score as a bounded scalar “ and persistent cache (claim 11). Each of the additional limitations is recited at a high level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component. Accordingly, these additional elements, alone or in combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
The claim does not include additional elements, alone or in combination, that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements, alone or in combination, are nothing more than mere instructions to apply the exception on a general computer.
Dependent claims 2-10 and 11-20 are also directed to an abstract idea without significantly more because they further narrow the abstract idea described in relation to claim 1/11 without successfully integrating the exception into a practical application (application tracking system is recited at a high level of generality and amounts to apply it instructions) or providing significantly more limitations.
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.
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.
Claim(s) 1-3, 10-13 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Mitchell (US 7593860) in view of Garg (US 10803421), Crow (WO 2005/038584) and Jain (US 20220171842).
As per claim 1/11, Mitchell discloses a system comprising:
a device associated with a prospective job candidate configured to provide a request for consideration for a first job type for the job candidate; a career management co-op platform (CMCP) system configured to:
receive the request (1:24-27, “(8) receiving by said computing apparatus, job candidate data, said job candidate data comprising a job candidate and a second list comprising a second plurality of required skills related to a job held by said job candidate;”);
determining one or more relevant skills possessed by the plurality of job candidates (3:49-53, “Job related data 16 comprises a list of specified job titles (e.g., computer programmers, data architect, network architects, etc) and associated required skills necessary to perform each specified job title.” 3:60-4:4, “A market valued skill is defined herein as a critical "hot" market place skill that is required to perform specified job and linked with business unit skills aligned to the achievement of business priorities (e.g., growth in revenue and innovation). A market valued skill comprises a job-related skill that when applied to a job will produce significant revenue, drive significant innovation, and is aligned with a market direction and particular business strategy for an entity or business. Market valued skills are dynamic and change as new products and services are introduced into the market place (e.g., as technology changes at a rapid pace market valued skills also change at a rapid pace).”);
Retrieving at least a portion of the aggregated job candidate data (claim 1, “wherein said first internal market data comprises first capacity planning data associated with an entity and first workforce management data associated with said entity,”, 4:25-30, “Internal market related data may comprise any market related data known to a person of ordinary skill in the art including, inter alia, resource and capacity planning data for an entity, workforce management data for an entity such as attrition and demand growth in specific environment areas (products/services/offerings), etc”);
A job-candidate match score is implemented as a bounded scalar with threshold comparison (6:45-52, “(18) Equation 1 will calculate analysis scores comprising a range of 0 to 2 with 0 indicating that the candidate is not a match for a specific job and 2 indicating that the candidate is a perfect match for a specific job.”, 6:49-59, “The analysis score is compared to a predetermined value to determine an action performed for the candidate (e.g., recommend that the candidate is a match for a specific job, reject a candidate for a specific job, recommend a training program to prepare the candidate for a specific job). For example, any analysis score comprising a range of greater than or equal to 1 (i.e., 1 is the predetermined value) is considered to be a good match for a specific job because the candidate is determined to possess an adequate number of required skills and therefore any training time required to move from a current job position to a desired job position would be minimal.”)
Comparing candidate’s current skills to required skills to produce a score (1:32-38, “(11) comparing by said career analysis tool, said first plurality of required skills with said second plurality of required skills to determine a first set of common skills between said first plurality of required skills and said second plurality of skills; (12) calculating, by said career analysis tool, a first score for said candidate by dividing the number of said first set of common skills by the number of said first plurality of required skills;”, 4:32-36, “(7) Career analysis tool 18 in system 2 of FIG. 1 allows job related data 16 to be compared to candidate input data 22 and a weighting factor(s) may be applied to the comparison to analyze a candidate's qualifications for specified job titles applied for by the candidate.”);
and based on the job seeker skillset match score satisfying a threshold, providing, to the device, a recommendation of a second job type relevant to at least one of the one or more relevant skills missing from the current skills of the prospective job candidate (6:60-63, “Any analysis score comprising a range of less than or equal to 1 (i.e., 1 is the predetermined value) is not considered to be a good match for a specific job because the candidate is not determined to possess an adequate number of required skills.”, 6:15-19, “If in step 44 it is determined that the analysis score is not greater than or equal to 1.0 (i.e., less than 1.0), then step 36 is repeated to perform another comparison between job related skills for a job title held by a candidate and required skills necessary for another desired job title.”, 7:43-48, “(19) Career analysis tool 18 may be used for (and the implementation example with respect to Tables 1-3 may be applied to) calculating several analysis scores for a candidate for several job titles and specific job titles may be recommended by career analysis tool 18 based on a comparison of each the analysis scores to each other.”, 6:45-7:4, “(18) Equation 1 will calculate analysis scores comprising a range of 0 to 2 with 0 indicating that the candidate is not a match for a specific job and 2 indicating that the candidate is a perfect match for a specific job. The analysis score is compared to a predetermined value to determine an action performed for the candidate (e.g., recommend that the candidate is a match for a specific job, reject a candidate for a specific job, recommend a training program to prepare the candidate for a specific job). For example, any analysis score comprising a range of greater than or equal to 1 (i.e., 1 is the predetermined value) is considered to be a good match for a specific job because the candidate is determined to possess an adequate number of required skills and therefore any training time required to move from a current job position to a desired job position would be minimal. Any analysis score comprising a range of less than or equal to 1 (i.e., 1 is the predetermined value) is not considered to be a good match for a specific job because the candidate is not determined to possess an adequate number of required skills. However, if there is no time limitation related to training time then candidates comprising an analysis score of less than 1 may be considered as long as required skills are determined to be considered market valued skills. Therefore, based of the implementation example with respect to Tables 1-3, the candidate was determined to have an analysis score of 1.3 (i.e., greater than or equal to 1) and therefore the candidate is considered to be a good match for the Network Architect job.”).
However, Mitchell does not disclose but Garg discloses retrieving, via an application programming interface (API), at least a portion of the aggregated job candidate data from an applicant tracking system (ATS) and/or a human resource information system (HRIS) (claim 2, “obtaining the talent profile from a human resource (HR) database of the first data source of the enterprise, wherein the talent profile of the candidate comprises at least one of: an employment role held by the candidate and skills associated with the role”, )
training, based on the aggregated job candidate data for the first job type and corresponding outcome labels comprising one or more of hiring outcomes and role-transition events, a machine learning model configured to output a job seeker skillset match score (claim 1, “a neural network module trained by adjusting at least one parameter associated with the neural network module based on a training enriched talent profile, a training calibrated job profile, and a training match score between the training enriched talent profile and the training calibrated job profile;”, 12:54-60, “(88) The system also has its own proprietary data 1090 with its history of matching enriched talent profiles to calibrated job positions across a variety of organizations. In certain embodiments, this data is used to train the DNN, obtain peer data, identify ideal and unqualified candidate for a calibrated job profile, and create vector representations of values for roles, skills, companies, schools, and fields of study.”, 13:10-16, “The matching engine 1040 matches enhanced talent profiles to calibrated job positions using DNN 1050 and outputs one or more hiring-related predictions for each candidate-job pair inputted into the engine. The match score calculation 1060 calculates match scores for candidate/job pairs using the hiring-related predictions outputting by the matching engine 1040.”);
determining, via the trained machine learning model based on a comparison of current skills possessed by the prospective job candidate and the relevant skills, the job seeker skillset match score (12:8-14, “The DNN 750 outputs one or more hiring-related predictions (980) for each pair, resulting in a match score for each of the inputted candidates at Organization X. Examples of hiring-related predictions are the probability of the candidate being interviewed, the probability of the candidate being hired, and the probability of the candidate being successful in a job after a period of time (e.g., 1 year).”, 13:10-16, “The matching engine 1040 matches enhanced talent profiles to calibrated job positions using DNN 1050 and outputs one or more hiring-related predictions for each candidate-job pair inputted into the engine. The match score calculation 1060 calculates match scores for candidate/job pairs using the hiring-related predictions outputting by the matching engine 1040.”))
Therefore, it would have been obvious to one of ordinary skill in the art at the time of the invention to include the limitation above as taught by Garg in the teaching of Mitchell, since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable.
However, Mitchell does not disclose but Crow discloses determining, based on aggregated job candidate data of a plurality of job candidates having the first job type, one or more relevant skills possessed by the plurality of job candidates ((page. 28, ln 20-31, “The method can be used to identify and extract skills from job candidate data (e.g., the job candidate data 122 of FIG. 1). At 1920, skills lists are identified, and at 1930, skills are extracted from the identified skills lists. The skills so extracted may then be added, for example, as skills with a confidence score. The confidence score can be compared with the confidence scores of the same concepts extracted by the other speculative extractors such as the other heuristic extractors or the parsing extractors. The confidence score for a particular concept can be added to the concept space responsive to determining that the confidence score reaches or exceeds the set threshold.”).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of the invention to include the limitation above as taught by Crow in the teaching of Mitchell, in order to find job candidate matches having characteristics similar to those of a desirable candidate (please see Crow abstract).
However, Mitchell does not disclose but Jain discloses generate a first temporary ID for the at least a portion of the aggregated job candidate data (claim 1, “generate, based on validation of the current credentials, a first credentials for the temporary identity, wherein the first credentials include the identifier of the temporary identity and the persistent source value; “, ); identify a first static ID associated with the prospective job candidate (abstract, “ A source value (e.g., an original identifier of an entity) is persisted across assumed identities, facilitating identification of entities (users or applications) responsible for actions taken by the assumed (e.g., alternative) identities. “); store, within a persistent cache, a record indicating a current association between the first temporary ID and the first static ID (claim 1, “wherein the first credentials include the identifier of the temporary identity and the persistent source value; and respond to the request with the first credentials that include the identifier of the temporary identity and the persistent source value copied from the request; “, abstract, “An Identity and Access Management Service implements persistent source values PSVs) for assumed identities.”).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of the invention to include the limitation above as taught by Jain in the teaching of Mitchell, in order to uncover an entire chain of temporary identities to trace an action back to the long-term identity responsible for the action(please see Jain Paragraph 3).
As per claim 2/12, Mitchell discloses wherein the CMCP system is further configured to, based on the job seeker skillset match score satisfying a threshold, determine the second job type relevant to the at least one of the one or more relevant skills missing from the current skills of the prospective job candidate (col. 6:45 to col. 7:15, “18) Equation 1 will calculate analysis scores comprising a range of 0 to 2 with 0 indicating that the candidate is not a match for a specific job and 2 indicating that the candidate is a perfect match for a specific job. The analysis score is compared to a predetermined value to determine an action performed for the candidate (e.g., recommend that the candidate is a match for a specific job, reject a candidate for a specific job, recommend a training program to prepare the candidate for a specific job). For example, any analysis score comprising a range of greater than or equal to 1 (i.e., 1 is the predetermined value) is considered to be a good match for a specific job because the candidate is determined to possess an adequate number of required skills and therefore any training time required to move from a current job position to a desired job position would be minimal. Any analysis score comprising a range of less than or equal to 1 (i.e., 1 is the predetermined value) is not considered to be a good match for a specific job because the candidate is not determined to possess an adequate number of required skills. However, if there is no time limitation related to training time then candidates comprising an analysis score of less than 1 may be considered as long as required skills are determined to be considered market valued skills. Therefore, based on the implementation example with respect to Tables 1-3, the candidate was determined to have an analysis score of 1.3 (i.e., greater than or equal to 1) and therefore the candidate is considered to be a good match for the Network Architect job. Additionally, prior to calculating the analysis score, a first fraction of common skills scores between the two job titles (i.e., Common skills*1/Total required skills from Equation 1) may be compared to a first predetermined value and a second fraction of market valued skills with respect to a total number of required skills for the requested job title (i.e., Weighting factor from Equation 1) may be compared to a second predetermined value. In this instance, the first predetermined value would have to be equal to or exceeded by the first fraction and the second predetermined value would have to be equal to or exceeded by the second fraction in order to calculate an analysis score.”). However, Mitchell does not disclose but Garg discloses using the trained machine learning model to generate a match score and determine appropriate job candidate pairings as disclosed in claim 1. (please see claim 1 rejection for combination rationale).
As per claim 3/13, Mitchell discloses determining the second job type based on a highest ranked skill of the one or more relevant skills missing from the current skills of the prospective job candidate (col. 7:29-42, “additionally, each of the market valued skills and each of the common skills may be weighted (i.e., a weighting factor applied) before an analysis score is calculated. As a first example, each of the common skills may be weighted according to a relative importance (e.g., by an expert in the field) to the requested job (e.g., for a Network Architect it may be determined that an Apply IT standards skill is more important that a communications skill and therefore the Apply IT standards skill is weighted higher). As a second example, each of the market valued skills may be weighted according to an amount of time that the market valued skill has been determined to be a market valued skill (e.g., the greater the amount of time that the market valued skill has been determined to be a market valued skill, the higher the weighting), col. 7:47-62, “Additionally, career analysis tool 18 may determine a skills delta between job related skills associated with a job held by the candidate and required skills necessary to obtain a desired position (i.e., required skills missing from the candidate's current job). Career analysis tool 18 may determine a training program for the candidate for obtaining the missing required skills determined by the skills delta. For example, with respect to the implementation example with respect to Tables 1-3, the candidate would be determined to have a skills delta of 3 (i.e., 3 missing required skills as indicated by "No" in Column 3, rows 3, 6, and 9 of Table 3). Career analysis tool 18 may develop a training program for the candidate so that the candidate could acquire the 3 missing required skills.” The highest ranked skill among the missing skills is simply the missing skill with the greatest weight). However, Mitchell does not disclose but Garg discloses using the trained machine learning model to generate a match score and determine appropriate job candidate pairings as disclosed in claim 1. (please see claim 1 rejection for combination rationale).
As per claim 10/20, Mitchell does not disclose but Crow discloses wherein the CMCP system is configured to receive aggregated job candidate data of a plurality of job candidates from an application tracking system (p.49:11-17, p.71:1-5, “The task of storing detailed information about a candidate can be left to the Applicant Tracking Software (ATS) that is the client of the Match Service. A set of CVOs can be returned from the match and clone methods of the Match Service API. It can also be the input to the clone method… Integration into Applicant Tracking Software System Any of the technologies described herein can be integrated into applicant tracking software system.”(please see claim 1 rejection for combination rationale).
Novelty and Non-Obviousness
No prior art was applied to claims 4-9 and 14-19 because the combination above with the prior art of record used to reject the claims in previous office actions will result in a piecemeal rejection using impermissible hindsight.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to OMAR ZEROUAL whose telephone number is (571)272-7255. The examiner can normally be reached Flex schedule.
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, Kevin Flynn can be reached at 5712703108. 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.
OMAR . ZEROUAL
Examiner
Art Unit 3628
/OMAR ZEROUAL/Primary Examiner, Art Unit 3629
Prosecution Insights