DETAILED ACTION
This nonfinal action is in response to application 18/603,381 filed on 03/13/2024.
Claims 1-20 are pending in the application. Claims 1, 8, and 15 are independent claims.
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 .
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the enablement requirement. The claims contain subject matter which was not described in the specification in such a way as to enable one skilled in the art to which it pertains, or with which it is most nearly connected, to make and/or use the invention.
Regarding, independent claims 1, 8, and 15, they recite the limitation "solving, by a processor set and using a machine learning algorithm based on Thistlethwaite's algorithm, the three-dimensional model to optimally match the one or more constraints with one or more strengths of respective one or more partners and one or more strengths of one or more connections between the respective one or more partners and respective one or more clients" (the 'solving ... to optimally match' limitation), and this limitation is inherited by every dependent claim (2-7, 9-14, and 16-20). The specification does not enable a person of ordinary skill in the art, a software engineer or data scientist familiar with machine learning and combinatorial optimization (BS/MS in computer science or similar), to make and use the full scope of this limitation without undue experimentation.
Whether undue experimentation is required is determined from the factors set forth in In re Wands, 858 F.2d 731, 737 (Fed. Cir. 1988), and MPEP 2164.01. Those factors, applied below, collectively establish that undue experimentation would be required to practice the full scope of the claimed 'solving ... to optimally match' limitation.
(A) Breadth of the claims: The 'solving ... to optimally match' limitation is recited in purely functional, result-oriented terms. It claims an end result, an 'optimal match' across an open-ended universe of partners, clients, connections, and project constraints (time, cost, scope; see [¶ 0066]), without limiting how the result is computed. The claimed genus thus encompasses every conceivable adaptation of Thistlethwaite's algorithm to every conceivable partnership-optimization problem, characterized only by what it accomplishes ('optimally match') rather than by any disclosed structure or operative steps. The enabling disclosure must be commensurate with this full functional scope (Amgen Inc. v. Sanofi, 598 U.S. 594, 610-14 (2023); Liebel-Flarsheim Co. v. Medrad, Inc., 481 F.3d 1371, 1379-80 (Fed. Cir. 2007)), and it is not. This factor weighs strongly against enablement.
(B) Nature of the invention: The asserted invention repurposes Thistlethwaite's algorithm, a deterministic, table-driven group-theory search developed to restore a scrambled Rubik's Cube to a single known solved state, as a 'machine learning algorithm' that 'optimally matches' partners, clients, connections, and project constraints. This is a cross-domain transplant of a closed-form combinatorial cube solver into an open-ended business-optimization domain, not an incremental improvement in a settled field. The specification itself flags the mismatch without resolving it: [¶ 0068] states the 'variation of Thistlethwaite's algorithm ... is an adjustment of Thistlethwaite's algorithm to conform to the constraints of the project, which is different from the color alignment constraint of a Rubik's Cube puzzle,' yet [¶ 0072]-[¶ 0077] describe only the original cube solver. Such a repurposing demands a disclosed bridge that the specification does not provide. This factor weighs against enablement.
(C) State of the prior art: Thistlethwaite's algorithm as a Rubik's Cube solver is well known, and [¶ 0072]-[¶ 0077] is a generic textbook recital of that known cube solver; general machine-learning libraries (TensorFlow, NumPy, Pandas; see [¶ 0071]) are likewise well known. What is NOT well known in the prior art is any technique for using Thistlethwaite's algorithm, or any 'variation' of it, as a 'machine learning algorithm' to solve an open-ended partnership/relationship optimization problem that has no single 'solved' state and no defined group or move set over partners, clients, and constraints. A person of ordinary skill in the art cannot draw on existing art to fill this gap because the art generally teaches the cube solver for the cube problem only. This factor weighs against enablement.
(D) Level of one of ordinary skill in the art: The person of ordinary skill in the art is a software engineer or data scientist with a BS/MS in computer science or similar, familiar with machine learning and combinatorial optimization. While a high level of skill ordinarily reduces the disclosure needed for routine implementation details, the gaps here are not routine details, they are the inventive core itself. The specification never defines the moves/generators, the G1-G4 analogues, the goal or 'solved' state representing an optimal business match, the objective function that defines 'optimal,' how partner/connection strengths and constraints are encoded into the matrix cells so that group operations are meaningful, or what makes the solver 'machine learning' (no training data, loss function, model, or training procedure is disclosed for the solver). Skill cannot substitute for an undisclosed algorithm. This factor does not save enablement.
(E) Level of predictability in the art: Although software is generally a predictable art for routine coding, predictability presupposes that the algorithm to be coded is disclosed; here it is not. There is no predictable path from a deterministic cube-solving group search to an open-ended 'optimal match' optimizer, and the specification supplies none. Whether any given 'adjustment' of Thistlethwaite's algorithm would converge on a meaningful 'optimal' partner-client-constraint match, or even be definable as a group operation over a non-cube state space, is unpredictable on this record, because the encoding of strengths/constraints into cells, the move set, the subgroup chain, and the goal state are all undefined ([¶ 0038], [¶ 0068]-[¶ 0069], [¶ 0071]). This factor weighs against enablement.
(F) Amount of direction provided by the inventor: The specification provides no operative direction for the claimed solver. Each passage touching the limitation merely restates the result or names tools: [¶ 0038] recites that the solver 'solve[s] the 3D matrix using a machine learning algorithm, which is a variation of Thistlethwaite's algorithm, to find an optimal solution' (conclusory); [¶ 0068] recites that a 'variation' is selected as an unspecified 'adjustment ... to conform to the constraints of the project' (stating THAT, never HOW); [¶ 0069] recites that the solver 'processes the 3D model to compute an optimal solution ... using the variation' (conclusory); and [¶ 0071] merely lists libraries (a 'variation of Thistlethwaite's algorithm,' NumPy, TensorFlow, Pandas). The only detailed algorithmic content, [¶ 0072]-[¶ 0077], describes the ORIGINAL cube solver and its G0->G1->G2->G3->G4 move restrictions, not any adaptation to partnership optimization or to machine learning. Merely stating that a computer performs a function without disclosing how fails enablement (Vasudevan Software, Inc. v. MicroStrategy, Inc., 782 F.3d 671, 681-82 (Fed. Cir. 2015)). This factor weighs heavily against enablement.
(G) Existence of working examples: The specification contains no working example of the claimed solving step and not even a prophetic algorithmic example. The 'EXAMPLES' at [¶ 0078]-[¶ 0079] are purely narrative business use-cases, a Partner Relations Manager visualizing and optimizing partner strengths for revenue sharing ([¶ 0078]) and an IT provider aligning its data-science strengths to client engagements ([¶ 0079]), and disclose no algorithmic inputs, no encoding of data into the 3D matrix, no move set, no objective function, no training procedure, and no solver outputs. Consistent with this, [¶ 0013] confirms the 3D model is merely 'reminiscent of a Rubik's Cube puzzle for ease of visualization and understanding', a metaphor, not an operative algorithm. This factor weighs against enablement.
(H) Quantity of experimentation needed to make or use the invention: Because the specification discloses only the desired result plus the original cube solver, a person of ordinary skill the art seeking to make and use the full scope of the 'solving ... to optimally match' limitation would have to independently invent every missing piece: a meaningful encoding of partner/connection strengths and project constraints into matrix cells; a defined group and move set/generators over the partner-client-constraint state space; nested-subgroup analogues to G1-G4; a goal or 'solved' state representing an open-ended 'optimal' business match (which, unlike the cube, has no single known solved state); an objective function defining 'optimal'; and a training data set, model, loss function, and training procedure to make the solver 'machine learning.' This is open-ended research to supply the inventive core across the entire claimed genus, not the routine experimentation the law tolerates. This factor weighs decisively against enablement.
Weighing the In re Wands factors collectively, the specification fails to enable a POSITA to make and use the full scope of the 'solving ... to optimally match' limitation without undue experimentation. The dispositive technical mismatch confirms the deficiency: Thistlethwaite's algorithm is a deterministic, table-driven group-theory search that restores a scrambled cube to ONE known solved state (the identity element H) via a fixed move set descending through nested subgroups G0..G4, whereas the claimed problem has no single known 'solved' state, no defined group or move set over partners-clients-constraints, and is not shown to be machine learning at all (no training data, loss function, model, or training procedure disclosed for the solver). The specification never bridges these gaps, it never defines the moves/generators, the G1-G4 analogues, the goal state representing an optimal match, the objective function defining 'optimal,' how strengths/constraints are encoded into the matrix cells so that group operations are meaningful, or what makes the method 'machine learning.' Because the scope of enablement is not commensurate with the scope of the claims, claims 1-20 are not enabled.
Dependent claims 6, 13, and 20 are even more directly directed to the undisclosed adaptation and are not cured by their additional limitations. Each recites 'selecting a variation of Thistlethwaite's algorithm based on the defined constraints, wherein the solving the three-dimensional model uses the selected variation of Thistlethwaite's algorithm', that is, they affirmatively claim the very 'variation'/'adjustment' of Thistlethwaite's algorithm that the specification states exists ([¶ 0068]) but never discloses how to create or apply. These claims therefore compound, rather than remedy, the enablement defect of the limitation from which they depend, and they remain rejected under 35 U.S.C. 112(a) for the same reasons.
Applicant is advised that a companion rejection under the written-description requirement of 35 U.S.C. 112(a) (functional claiming of a result without disclosing how it is achieved; see Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co., 598 F.3d 1336 (Fed. Cir. 2010) (en banc); Vasudevan Software, Inc. v. MicroStrategy, Inc., 782 F.3d 671 (Fed. Cir. 2015)) is also applicable and reinforces this enablement rejection.
Regarding, dependent claims 2-7, 9-14, and 16-20, they do not cure the 35 U.S.C. 112(a) enablement and written-description deficiency identified in independent claims 1, 8, and 15. Each dependent inherits the un-enabled and inadequately-described limitation of 'solving ... using a machine learning algorithm based on Thistlethwaite's algorithm ... to optimally match' the constraints with partner and connection strengths, and none supplies the missing disclosure of how that solving is actually performed, e.g. the move set/generators over partners-clients-constraints, the subgroup chain or 'solved'/goal state representing an optimal business match, the objective function defining 'optimal,' the encoding of strengths and constraints into the matrix cells, or any training data, loss function, model, or training procedure that would render the solver 'machine learning.' The added limitations instead elaborate only the separately-disclosed upstream profiling (claims 2-3, 9-10, 16-17), relationship-graph (claims 4-5, 11-12, 18-19), matrix-initialization (claims 6, 13, 20), and recalibration (claims 7, 14) steps. Claims 6, 13, and 20 are even more directly implicated: they affirmatively recite 'select[ing] a variation of Thistlethwaite's algorithm based on the defined constraints' and use that selected variation in the solving step, yet the specification states only conclusorily that such a 'variation' or 'adjustment to conform to the constraints of the project' exists ([¶ 0038], [¶ 0068]-[¶ 0069]) and names software libraries ([¶ 0071]) without disclosing how any variation is defined, selected, or applied, and the algorithm overview at paragraphs [¶ 0072]-[¶ 0077] describes only the original Rubik's-Cube solver. Accordingly, claims 6, 13, and 20 aggravate rather than remedy the defect. The dependent claims are therefore rejected under 35 U.S.C. 112(a) for the same reasons as their respective independent claims.
Claims 1-20 are rejected under 35 U.S.C. 112(a) as failing to comply with the written description requirement. The claims contain subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor, at the time the application was filed, had possession of the claimed invention.
Independent claims 1, 8, and 15 each recite "solving, by a processor set and using a machine learning algorithm based on Thistlethwaite's algorithm, the three-dimensional model to optimally match the one or more constraints with one or more strengths of respective one or more partners and one or more strengths of one or more connections between the respective one or more partners and respective one or more clients." This limitation is a recitation of a desired result (an optimal match) achieved by a solver, but the specification does not describe how that result is achieved. Under Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co., 598 F.3d 1336 (Fed. Cir. 2010) (en banc), a specification must demonstrate possession of the claimed invention, and where a claim recites a function or result, the specification must disclose the structure, species, or algorithm that performs it rather than merely the result desired. For computer-implemented functional limitations, Vasudevan Software, Inc. v. MicroStrategy, Inc., 782 F.3d 671 (Fed. Cir. 2015), requires that the specification disclose the algorithm by which the claimed computer function is performed; a disclosure that the computer performs the function, without the algorithm, does not show possession. See MPEP 2161.01 and MPEP 2163.
The specification here does not disclose any algorithm by which the recited 'optimally match' function is performed. First, the specification discloses no objective function: it never defines, in computational terms, what 'optimal' means, never identifies a quantity to be minimized or maximized, and never sets out a scoring or decision criterion for selecting among candidate matches. The disclosure instead restates the claimed result in conclusory terms, see [¶ 0038] ('to find an optimal solution that offers an optimal combination'), [¶ 0042]-[¶ 0043] (FIG. 3, steps 306/308, 'to optimally match'), and [¶ 0068]-[¶ 0069] (FIG. 7, steps 706/708, 'compute an optimal solution ... using the variation'). Second, the specification does not map Thistlethwaite's algorithm to the claimed problem domain. Thistlethwaite's algorithm, as the specification itself describes at [¶ 0072]-[¶ 0077], is a method for solving a Rubik's Cube that operates over the cube's standard move set and progresses through nested subgroups G0 -> G1 -> G2 -> G3 -> G4 to a single solved state represented by the identity element H. The specification provides this generic textbook overview of the original cube algorithm but never discloses the analogous move set or generators over partners, clients, and constraints; never discloses analogues of the G1-G4 subgroups; never identifies a 'solved' or goal state representing an optimal partnership match (and, unlike a cube, the recited problem has no single known solved state); and never discloses how partner strengths and project constraints are encoded into the cells of the three-dimensional matrix so that such group operations would be defined or meaningful. [¶ 0068] states only that the 'variation of Thistlethwaite's algorithm is an adjustment ... to conform to the constraints of the project, which is different from the color alignment constraint of a Rubik's Cube,' but discloses only THAT an adjustment exists and never discloses HOW the adjustment is made. Third, although the claims and specification characterize the solver as 'a machine learning algorithm,' the specification discloses no machine-learning scheme for the solver, no training data, no loss or objective function, no model architecture, and no training procedure. [¶ 0071] names software libraries (a NumPy library, a TensorFlow library 'for implementing the machine learning algorithm,' and a Pandas library), but naming a library identifies a tool, not the algorithm performing the claimed function. The 'EXAMPLES' at [¶ 0078]-[¶ 0079] are narrative business use cases that disclose no algorithmic inputs, encodings, steps, or outputs.
Because the specification recites only the result of the solving step and a black-box solver, without disclosing the objective function defining 'optimal,' the mapping of Thistlethwaite's moves/subgroups/solved-state to the partnership domain, or any training scheme that would make the solver machine learning, it does not reasonably convey that the inventor possessed the claimed solving function at the time of filing. Independent claims 1, 8, and 15 functionally encompass the entire genus of 'machine learning algorithm[s] based on Thistlethwaite's algorithm' that produce the recited optimal match, yet the specification describes not a single working species and none of the algorithmic features common to the genus. Under Ariad, describing only the desired result of a genus, without a representative number of species or the common structural/algorithmic features, does not demonstrate possession. Independent claims 1, 8, and 15 are therefore rejected.
Dependent claims 6, 13, and 20, which recite 'selecting a variation of Thistlethwaite's algorithm based on the defined constraints, wherein the solving the three-dimensional model uses the selected variation of Thistlethwaite's algorithm,' are rejected for the same reasons and are more directly implicated, because they presuppose a described family of constraint-specific variations and a criterion for selecting among them, yet the specification discloses no such variation and no selection criteria mapping particular constraints to particular variations. Dependent claims 2-5, 7, 9-12, 14, and 16-19 are rejected because they depend from claims 1, 8, or 15, inherit the inadequately described solving limitation, and do not add any algorithmic disclosure that cures the deficiency. Claims 1-20 are accordingly rejected under 35 U.S.C. 112(a) for failure to satisfy the written description requirement. This written description rejection is a companion to the enablement rejection of the same limitation and is set forth as an independent and additional ground.
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 an abstract idea without significantly more. The analysis of the claims will follow the 2019 Revised Patent Subject Matter Eligibility Guidance, 84 Fed. Reg. 50 (“2019 PEG”).
Independent Claims (Claim 1, Claim 8, Claim 15):
Step 1: Claim 1 is drawn to a method, claim 8 is drawn to a system/apparatus, and claim 15 is drawn to a product. Therefore, each of these claims falls under one of the four categories of statutory subject matter (process/method, machine/apparatus, manufacture/product, or composition of matter).
Step 2A Prong 1: Claims 1, 8, and 15 each recite a judicially recognized exception of an abstract idea.
Claim 1 recites, inter alia:
generating a categorized profile of partners based on feedback from clients and past performances of the partners, the categorized profile indicating strengths of the partners; generating a network graph based on data about interactions between the partners and the clients and further based on the categorized profile of the partners, the network graph having nodes representing the partners and the clients and having edges representing connections between the partners and the clients – These limitations recite certain methods of organizing human activity, namely commercial interactions and the management of business relationships and sales: it profiles business partners by strength based on client feedback and past performance, maps partner-client relationships, and matches partners to project constraints associated with commercial outcomes. This is a fundamental commercial/business-management practice/managing relationships between people that falls squarely within MPEP 2106.04(a)(2)(II).
generating, using the categorized profile and the network graph, a three-dimensional model represented by a three-dimensional matrix of cells, a given cell representing a part of a project and including one or more constraints of the project: These limitations recite mathematical relationships similar to organizing information and manipulating information, e.g. generating a three-dimensional model represented by a three-dimensional matrix of cells, a given cell representing a part of a project and including one or more constraints of the project, through mathematical correlations, e.g. using the categorized profile and the network graph, see MPEP 2106.04(a)(2)(I)(A)(iv).
and solving, based on Thistlethwaite's algorithm, the three-dimensional model to optimally match the one or more constraints with one or more strengths of respective one or more partners and one or more strengths of one or more connections between the respective one or more partners and respective one or more clients - These limitations name a specific solving/optimization algorithm and recites the mathematical concept of computing an optimal solution over a matrix; this is analogous to Example 47 claim 2 (naming backpropagation/gradient descent), and recites, not merely involves, a mathematical concept.
Claims 8 and 15 recite substantially similar abstract idea limitations to those found in claim 1, and therefore recite the same judicial exception.
Step 2A Prong 2: The following additional elements recited in claims 1, 8, and 15 also do not integrate the recited judicial exceptions into a practical application.
Claim 1 additionally recites:
A computer-implemented method / [solving] by a processor set – These recite generic computer hardware and a generic computing environment performing generic functions. This is the classic 'apply it on a computer' scenario (MPEP 2106.05(f)) and a mere field-of-use/technological-environment link (MPEP 2106.05(h)); no particular machine (2106.05(b)) and no transformation (2106.05(c)) is present.
[generating] using a supervised machine learning model / [solving] by a machine learning algorithm – These are recited generically as tools to implement abstract process steps. Per the August 2025 Kim memo and MPEP 2106.05(f), merely applying a generic ML model / algorithm to carry out the abstract idea is 'apply it' and does not integrate the exception. The claim recites no improvement to the ML model / algorithm itself (Recentive Analytics v. Fox).
Claims 8 and 15 recite substantially similar additional elements to those found in claim 1, and therefore also do not integrate the recite judicial exceptions into a practical application.
Step 2B: The additional elements recited in claims 1-20, viewed individually or as an ordered combination, do not provide an inventive concept or otherwise amount to significantly more than the recited abstract ideas themselves.
Claim 1 additionally recites:
A computer-implemented method / [solving] by a processor set – Mere instructions to implement an abstract idea on a computer or computer components do not provide an inventive concept or significantly more to the recited abstract idea.
[generating] using a supervised machine learning model / [solving] by a machine learning algorithm – Merely applying a generic ML model / algorithm to carry out abstract process steps does not provide an inventive concept or significantly more to the recited abstract idea.
Claims 8 and 15 recite substantially similar additional elements to those found in claim 1, and therefore also do not provide an inventive concept or significantly more to the recited abstract idea.
Even when considered as an ordered combination, the additional elements are recited at a high level of generality and amount to no more than instructions to apply the abstract idea on a generic computer using generic data structures and algorithms. The asserted benefit is a better business outcome (better partner matches yielding more sales/revenue), which is an improvement to the abstract idea/business method itself, not an improvement to computer functionality or any technology. None of the additional elements imposes a meaningful limit that transforms the abstract idea into patent-eligible subject matter.
As such, claims 1, 8, and 15 are not patent eligible.
Dependent Claims (Claims 2-7, Claims 9-14, Claims 16-20):
Dependent claims 2-7, 9-14, and 16-20 narrow the scope of independent claims 1, 8, and 15, and likewise narrow the recited judicial exceptions. They recite abstract idea limitations that are similar to those recited within the independent claims (i.e., mental processes and/or mathematical concepts), and thereby merely expand on the already recited exceptions. The dependent claims also do not recite any further additional elements that successfully integrate the recited judicial exceptions into a practical application or provide significantly more than the recited abstract ideas themselves. Consequently, claims 2-7, 9-14, and 16-20 are also rejected under 35 U.S.C. 101.
Step 1: Claims 2-7 are drawn to a method, claim 9-14 are drawn to a system/apparatus, and claims 16-20 are drawn to a product. Therefore, each of these claims falls under one of the four categories of statutory subject matter (process/method, machine/apparatus, manufacture/product, or composition of matter).
Step 2A Prong 1: Claims 2-7, 9-14, and 16-20 each recite a judicially recognized exception of an abstract idea.
Claim 2 recites, inter alia:
wherein the generating the categorized profile includes: collecting, cleaning, and normalizing data about the partners by using principal component analysis to select features; extracting the features by using principal component analysis (PCA); training the supervised learning model by using the identified and extracted features and a k-Nearest Neighbors algorithm: – These limitations include 'principal component analysis (PCA)' and a 'k-Nearest Neighbors algorithm,' which are named mathematical/statistical techniques and therefore recite mathematical concepts. Under the August 2025 Kim memo recites-vs-merely-involves discipline, PCA and k-NN are expressly named calculations and are properly treated within the mathematical-concepts grouping
the cleaning and the normalizing providing a consistency and an accuracy in the collected data, and the collected data including areas of expertise of the partners, past performances of the partners, and client feedback about the partners; identifying features in the collected, cleaned, and normalized data having a relevancy to the strengths of the partners, the features including knowledge areas of the partners, experience levels of the partners, and a history of relationships involving the partners – These limitations recite acts that continue and elaborate the same abstract idea identified in claim 1, a method of organizing human activity in the form of commercial interactions and managing business relationships by evaluating partners' strengths from client feedback and past performance. The same acts could be performed by a person (a manager) collecting and reviewing partner information and judging each partner's strengths
Claim 3 recites, inter alia:
continuously updating and refining the collected data based on new data about the areas of expertise, the past performances, and the client feedback to maintain an accuracy in a categorization of the partners based on the strengths of the partners – These limitations recite further abstract ideas of profiling business partners by strength from client feedback and past performance and managing those business relationships, a certain method of organizing human activity (commercial interactions / managing relationships). It also amounts to a generic recalculation/iteration over the same data, a mathematical concept. A business manager can, as a mental and managerial process, periodically revisit and refine partner evaluations as new feedback and performance information arrives.
Claim 4 recites, inter alia:
collecting data about the interactions between the partners and the clients, the collected data including a frequency of the interactions, rates of success resulting from the interactions, and scores indicating feedback from the clients and the partners about the interactions; cleaning the collected data to remove outliers, manage missing values, and validate relationship metrics included in the collected data – These limitations recite how a person managing business development would gather feedback on client relationships, judge which partner-client relationships are strong and trustworthy, and rank them - activity that can be performed mentally or with pen and paper and that constitutes organizing human commercial activity.
wherein the generating the network graph is based on the cleaned data; estimating, using a graph database model, strengths of the connections between the partners and the clients by measuring weights of the edges in the network graph and using centrality measures, the weights being based on past interactions between the partners and the clients, the strengths of the partners, and measures of trustworthiness of the partners and the clients – These limitations further recite mathematical relations similar to organizing information and manipulating information, e.g. wherein the generating the network graph is based on the cleaned data; estimating, using a graph database model, strengths of the connections between the partners and the clients, through mathematical correlations, e.g. by measuring weights of the edges in the network graph and using centrality measures, the weights being based on past interactions between the partners and the clients, the strengths of the partners, and measures of trustworthiness of the partners and the clients, see MPEP 2106.04(a)(2)(I)(A)(iv).
and generating, using the graph database model, a client-partner relationship index based on the estimated strengths of the connections between the partners and the clients – These limitations further recite mathematical relations similar to organizing information and manipulating information, e.g. and generating, using the graph database model, a client-partner relationship index, through mathematical correlations, e.g. based on the estimated strengths of the connections between the partners and the clients, see MPEP 2106.04(a)(2)(I)(A)(iv).
Claim 5 recites, inter alia:
continuously updating the client-partner relationship index and the network graph by using data about new interactions between the partners and the clients to re-estimate the strengths of the connections between the partners and the clients – These limitations recite re-evaluating the quality/strength of business relationships from new interaction data is a commercial/managerial activity that falls within the 'certain methods of organizing human activity' grouping (managing business relationships and commercial interactions).
Claim 6 recites, inter alia:
wherein the generating the three-dimensional matrix of cells includes: defining constraints of the project by using the categorized profile of the partners and the network graph, the constraints including a time, a cost, and a scope of the project; initializing the three-dimensional model as the three-dimensional matrix – These limitations further recite mathematical relations similar to organizing information and manipulating information, e.g. wherein the generating the three-dimensional matrix of cells includes: defining constraints of the project, through mathematical correlations, e.g. by using the categorized profile of the partners and the network graph, the constraints including a time, a cost, and a scope of the project; initializing the three-dimensional model as the three-dimensional matrix, see MPEP 2106.04(a)(2)(I)(A)(iv).
Claim 7 recites, inter alia:
continuously updating the three-dimensional matrix based on data about new projects and alterations in constraints of existing projects – These limitations further maintaining and re-deriving the partner-to-project match as new projects arrive and as project constraints change is part of managing commercial interactions and business relationships to optimize sales/revenue (certain methods of organizing human activity).
Claims 9-14 and 16-20 recite substantially similar abstract idea limitations to those found in claims 2-7, and therefore recite the same judicial exceptions.
Step 2A Prong 2: Claims 4-5, 7, 11-12, 14, and 18-19 do not recite any further additional elements beyond those already recited in the independent claims. The following additional elements recited in claims 2-3, 6, 9-10, 13, 16-17, and 20 also do not integrate the recited judicial exceptions into a practical application.
Claim 2 additionally recites:
[using principal component analysis] and the supervised learning model to select features; [identifying] by the supervised learning model [features] – These are recited generically as a tool to implement abstract process steps. Per the August 2025 Kim memo and MPEP 2106.05(f), merely applying a generic ML model / algorithm to carry out the abstract idea is 'apply it' and does not integrate the exception. The claim recites no improvement to the ML model / algorithm itself (Recentive Analytics v. Fox).
validating the trained supervised learning model by using a separate data set to ensure a reliability in predictions by the supervised learning model – These additional elements recite generic, routine machine-learning practice (train/validate split) recited at a high level of generality. It is insignificant extra-solution activity, adding no practical application see MPEP 2106.05(g).
Claim 3 additionally recites:
[updating] the supervised learning model based on new data – This additional element is recited at a high level of generality with no details of how the supervised learning model is architected/refined other than reciting a result and thus cannot integrate the judicial exception into a practical idea see MPEP 2106.05(f). Under the August 2025 Kim memo, merely applying or retraining a generic ML model on a judicial exception does not integrate the exception into a practical application; the claim recites no improvement to the model's architecture or training process. This is mere instruction to apply the exception using generic ML on a generic computer (MPEP 2106.05(f)) and links the exception to a technological environment (MPEP 2106.05(h)).
Claim 6 additionally recites:
and selecting a variation of Thistlethwaite's algorithm based on the defined constraints, wherein the solving the three-dimensional model uses the selected variation of Thistlethwaite's algorithm – These additional elements are recited at a high level of generality with no improvement to the algorithm itself (Kim memo). This is 'apply it' on a generic computer (MPEP 2106.05(f)) and linking to a technological environment (MPEP 2106.05(h)
Claims 9-10, 13, 16-17, and 20 recite substantially similar additional elements to those found in claims 2-3 and 6, and therefore also do not integrate the recited judicial exceptions into a practical application.
Step 2B: The additional elements recited in claims 2-3, 6, 9-10, 13, 16-17, and 20, viewed individually or as an ordered combination, do not provide an inventive concept or otherwise amount to significantly more than the recited abstract ideas themselves.
Claim 2 additionally recites:
[using principal component analysis] and the supervised learning model to select features; [identifying] by the supervised learning model [features] – Merely applying a generic ML model / algorithm as a tool to carry out an abstract idea does not provide an inventive concept or significantly more to the recited abstract idea.
validating the trained supervised learning model by using a separate data set to ensure a reliability in predictions by the supervised learning model – Model validation is well-understood, routine, and conventional activity (not that specification admits use of conventional ML tooling, scikit-learn at [¶ 0057] and NumPy/TensorFlow at [¶ 0071], used in conventional ways to train and validate models), and therefore does not provide an inventive concept or significantly more to the recited abstract idea.
Claim 3 additionally recites:
[updating] the supervised learning model based on new data – Merely applying or retraining a generic ML model to carry out an abstract idea does not provide an inventive concept or significantly more to the recited abstract idea.
Claim 6 additionally recites:
and selecting a variation of Thistlethwaite's algorithm based on the defined constraints, wherein the solving the three-dimensional model uses the selected variation of Thistlethwaite's algorithm – Additional elements recited at a high level of generality with no improvement to the algorithm itself (Kim memo), and which are 'apply it' on a generic computer (MPEP 2106.05(f)) and linking to a technological environment (MPEP 2106.05(h)), do not provide an inventive concept or significantly more to the recited abstract idea.
Claims 9-10, 13, 16-17, and 20 recite substantially similar additional elements to those found in claims 2-3 and 6, and therefore also do not provide an inventive concept or significantly more to the recited abstract idea.
As such, claims 2-7, 9-14, and 16-20 also are not patent eligible.
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.
Claims 1, 6-7, 8, 13-14, 15, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar et al. (Pub. No. US 20240386316 A1, “Generating Opportunity Profile Insights”, filed 05/17/2023), hereinafter Kumar, in view of Tamayo (Pub. No. US 20180075126 A1, “Time-sensitive Cube”, published 03/25/2018) and Chen (“Different Algorithms to Solve a Rubik’s Cube”, published 2022).
Regarding claim 1, Kumar teaches A computer-implemented method comprising:
generating, using a supervised machine learning model, a categorized profile of partners (“The one or more clustering models 128 utilized by the clustering engine 120 may be configured to group historical opportunity profiles 108 into one or more opportunity profile cluster 110 based on one or more similarity or distance measures without pre-existing labels or categories for the historical opportunity profiles 108. The one or more clustering models 128 may include one or more machine learning models, such as one or more unsupervised machine learning models, supervised machine learning models, or one or more semi-supervised machine learning models” [Kumar ¶ 0062]) based on feedback from clients and past performances of the partners, the categorized profile indicating strengths of the partners (“The plurality of opportunity parameters may include at least one of: transaction parameters, segmentation parameters, personnel parameters, or prospect parameters. An outcome indicator associated with a historical opportunity profile 108 may indicate, with respect to the historical opportunity, one or more characteristics pertaining to at least one of: an outcome, a result, a reaction, an effect, a consequence, a decision, an aftermath, a conclusion, a materialization” [Kumar ¶ 0030]; “In one example, an outcome indicator may represent an outcome based on a performance metric. The indication of the outcome may include at least one of: a value for the performance metric, an indication as to whether the value for the performance metric meets a threshold (e.g., a goal or a benchmark), or an indication as to a range (such as from a set of ranges) the value for the performance metric falls within... the performance metric may include one or more of: total sales value, return on investment (ROI), return on equity (ROE), return on assets (ROA), return on sales (ROS), return on marketing investment (ROMI), gross profit margin (GPM), customer acquisition cost (CAC), customer lifetime value (CLV), sales cycle length, time to close, or the like” [Kumar ¶ 0051]; “The historical opportunity profiles 108 in an opportunity profile cluster 110 may be partitioned into subsets 114 (e.g., subset 114a and subset 114b). In one example, the subsets 114 may be partitioned from one another based on one or more outcome indicators 112. The subsets 114 may represent similarities, differences, patterns, relationships, or the like, that are based on the respective outcome indicator 112. For example, outcome indictors 112c and 112d may represent one kind of outcome, and outcome indicators 112e and 112f may represent another kind of outcome. In one example, outcome indicators 112c and 112d may indicate a favorable outcome, such as having won a potential prospect, and outcome indicators 112e and 112f may represent an unfavorable outcome, such as having lost a potential prospect.” [Kumar ¶ 0058]);
generating a network graph based on data about interactions between the partners and the clients and further based on the categorized profile of the partners, the network graph having nodes representing the partners and the clients and having edges representing connections between the partners and the clients (“In one example, a clustering model may include at least one of:..a graph-based model” [Kumar ¶ 0065]; “A graph-based model may include one or more components that represent data as a graph that includes a collection of nodes and edges that connect the nodes. The nodes may represent an object or entity, and the edges may represent a relationship or interaction between the objects. An example graph-based model may include a graph neural network, a random walk model, a signed graph model, or a community detection model” [Kumar ¶ 0070]).
However, Kumar does not expressly teach generating, using the categorized profile and the network graph, a three-dimensional model represented by a three-dimensional matrix of cells, a given cell representing a part of a project and including one or more constraints of the project.
In the same field of endeavor, Tamayo teaches a means of modeling a multidimensional database that generat[es] a three-dimensional model (According to an embodiment, a computer-implemented method of providing multi-dimensional time series objects to a user is disclosed that may comprise: providing an electronic database configured to store a plurality of time-series objects including one or more time-series metric objects and a plurality of dimension objects; generating, by a computing system including one or more hardware computer processors, based at least in part on the plurality of time-series objects, a time-sensitive OLAP cube” [Tamayo ¶ 0006]) represented by a three-dimensional matrix of cells, a given cell representing a part of a project and including one or more constraints of the project (“Online Analytical Processing (OLAP) Cube: A group of data cells and/or database items arranged according to the dimensions of the data. When the data includes three or more dimensions, the data may be visualized as a cube or hypercube in which each dimension forms a side of the cube. Example dimensions may include measures, metrics, products, geographical regions, and sales channels, among others. The data of an OLAP cube is organized such that the OLAP cube may be manipulated and operated upon in various ways such that a user may rapidly extract relevant data” [Tamayo ¶ 0054]; “Advantageously, each cell of the time-sensitive OLAP cube 550 may represent a time-series object. When operations are performed on the time-sensitive OLAP cube 550, one or more time-series objects are output, which may then be further visualized and analyzed. Generally, time-sensitive OLAP cubes may have hierarchies or formula-based relationships of data within each dimension” [Tamayo ¶ 0079])
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated generating a three-dimensional model represented by a three-dimensional matrix of cells, a given cell representing a part of a project and including one or more constraints of the project as taught by Tamayo into Kumar because they are both directed towards means of modeling and interpreting multidimensional data. Incorporating the teachings of Tamayo would enable faster querying of the data and further support end user analysis (“Advantageously, the multidimensional database structure 300 is organized such that it may be used in the time-sensitive cube data system and enable rapid responses to multidimensional queries and operations. Further, the responses and/or outputs of queries to the time-sensitive cube data system may include time-series objects, to which further expressions and/or statistical analysis may be applied. Thus, time-sensitive metrics and/or statistics may be extracted from the time-sensitive cube data system” [Tamayo ¶ 0073]).
However, the combination of Kumar and Tamayo does not expressly teach solving, by a processor set and using a machine learning algorithm based on Thistlethwaite' s algorithm, the three-dimensional model to optimally match the one or more constraints with one or more strengths of respective one or more partners and one or more strengths of one or more connections between the respective one or more partners and respective one or more clients,
In the same field of endeavor, Chen teaches a means of optimally solving a bounded constraint space problem through machine learning, wherein the disclosed machine learning algorithm is based on Thistlethwaite's algorithm (“This essay introduces and compares four methods to solve a Rubik’s cube which are Korf’s algorithm, Thistlethwaite’s algorithm, Kociemba’s algorithm and machine learning method” [Chen Abstract]; “Thistlethwaite’s algorithm is to restrict the moves in several stages in order to reduce the state space, or in other words, to prune the branches that is undesirable” [Chen page 4 Thistlethwaite’s Algorithm]; “First, the model is given a cube in its initial scrambled state. Then, by applying all the moves from M, the author can obtain the first mutated population. Now, by using the trained model, it can estimate the best move using policy p and value v. When the cube is applied with that move, it will get closer to the solved state. Then, repeating the steps until the solution is found” [Chen page 7 Applying the model]; The disclosed machine learning algorithm and Thistlethwaite’s algorithm rely on similar structural logic of gradually reducing the branching factor of a massive search space to find an optimal path) .
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the machine learning algorithm based on Thistlethwaite’s algorithm taught by Chen into Kumar and Tamayo because both Tamayo and Chen are directed towards modeling of bounded constraint space. Incorporating the machine learning algorithm of Chen would allow one of ordinary skill in the art to efficiently navigate the complex, multi-variable constraint space defined by Tamayo and arrive at an optimal resource match (“Rubik’s cube is very complex and has substantial number of possibilities, which makes it extra sophisticate to solve. By applying algorithms to deal with this problem, the potential of using computers and algorithms to deal with problems with this level of complexity can be seen. Therefore, the author can see that it has the potential to solve problems in other areas that are significant to our society” [Chen Abstract]).
Regarding claim 6, the combination of Kumar, Tamayo, and Chen teaches the limitations of parent claim 1, and Tamayo further teaches defining constraints of the project by using the categorized profile of the partners and the network graph, the constraints including a time, a cost, and a scope of the project (“an electronic database configured to store a plurality of time-series objects including one or more time-series metric objects and a plurality of dimension objects” [Tamayo ¶ 0006]; “In an embodiment, dimensions 306 and 310 may include more and/or fewer characteristics than is shown in FIG. 3. In an embodiment, the dimension characteristics 308 and 310 may be organized in a hierarchical structure, including, for example, sub-characteristics, sub-sub-characteristics, etc. In an embodiment, dimension 306 and/or dimension 310 may include objects, time-series objects, and/or other types of data or labels. In an embodiment, characteristics of a dimension may be referred to as values” [Tamayo ¶ 0072]; “The visual, cube-like, representation of FIG. 5 is referred to as a time-sensitive OLAP cube 550. Three dimensions of data are represented in the time-sensitive OLAP cube 550: servicer 410, status 406, and state 552. The state dimension 552 includes characteristics CA (California), FL (Florida), and IL (Illinois), while the servicer 410 and status 406 dimensions include values as described above with reference to FIG. 4” [Tamayo ¶ 0077]; Dimensions of the OLAP cube (i.e., three-dimensional model) are broadly definable under any characteristics defining the state space, e.g., time, cost, scope constraints (i.e., opportunity parameters as defined in Kumar [see Kumar ¶ 0031] );
initializing the three-dimensional model as the three-dimensional matrix; ([Tamayo ¶ 0006, ¶ 0054] as detailed in claim 1 above)
Chen further teaches selecting a variation of Thistlethwaite' s algorithm based on the defined constraints, wherein the solving the three-dimensional model uses the selected variation of Thistlethwaite' s algorithm ([Chen Abstract and page 7 Applying the model] as detailed in claim 1 above).
Regarding claim 7, the combination of Kumar, Tamayo, and Chen teaches the limitations of parent claim 1, and Tamayo further teaches continuously updating the three-dimensional matrix based on data about new projects and alterations in constraints of existing projects (“the time-sensitive OLAP cube 550 of FIG. 5 includes data for a single point in time (a snapshot), but may be slid over time (e.g., updated over time) to include data for any point in time” [Tamayo ¶ 0090])
Regarding claims 8 and 13-14, they are apparatus claims that substantially correspond to the methods of claims 1 and 6-7, which are already taught by the combination of Kumar, Tamayo, and Chen as detailed above. Kumar further teaches A computer system comprising: a processor set; a set of one or more computer-readable storage media; and program instructions, collectively stored in the set of one or more computer-readable storage media, for causing the processor set to perform the claimed computer operations ([Kumar ¶ 0138-0140]). Consequently, claims 8 and 13-14 are rejected for the same reasons as claims 1 and 6-7.
Regarding claims 15 and 20, they are product claims that substantially correspond to the methods of claims 1 and 6, which are already taught by the combination of Kumar, Tamayo, and Chen as detailed above. Kumar further teaches A computer program product comprising: a set of one or more computer-readable storage media; and program instructions, collectively stored in the set of one or more computer-readable storage media, for causing a processor set to perform the claimed computer operations ([Kumar ¶ 0138-0140]). Consequently, claims 15 and 20 are rejected for the same reasons as claims 1 and 6.
Claims 2-3, 9-10, and 16-17 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar, Tamayo, and Chen, as applied to claims 1, 8, and 15 above, further in view of LaViale (“Deep Dive on KNN: Understanding and Implementing the K-Nearest Neighbors Algorithm”, published online 03/16/2023).
Regarding claim 2, the combination of Kumar, Tamayo, and Chen teaches the limitations of parent claim 1, and Kumar further teaches collecting, cleaning, and normalizing data about the partners by using principal component analysis and the supervised learning model to select features, the cleaning and the normalizing providing a consistency and an accuracy in the collected data, (“In at least one example, the matching module 122 may scale or normalize the data represented by the opportunity profile clusters 110…Additionally, or in the alternative, one or more dimensionality reduction operations may be performed, for example, upon one or more high-dimensional aspects of the data represented by the opportunity profile clusters 110. The dimensionality reduction operations may include principal component analysis (PCA) or t-distributed stochastic neighbor embedding (t-SNE)” [Kumar ¶ 0073]) and the collected data including areas of expertise of the partners, past performances of the partners, and client feedback about the partners; (see parameters and outcome indicators [Kumar ¶ 0030, 0051] as detailed in claim 1 above)
identifying, by the supervised learning model, features in the collected, cleaned, and normalized data having a relevancy to the strengths of the partners, the features including knowledge areas of the partners, experience levels of the partners, and a history of relationships involving the partners; (“In at least one example, the matching criteria 130 may include selecting a subset of opportunity parameters for comparison of the target opportunity profile 116 to the opportunity profile clusters 110. For example, the subset may include a subset of the transaction parameters, a subset of the personnel parameters, or a subset of the prospect parameters” [Kumar ¶ 0074]; also see opportunity parameters [Kumar ¶ 0030] as detailed in claim 1 above)
extracting the features by using principal component analysis (PCA); ([Kumar ¶ 0073] as detailed above)
training the supervised learning model by using the identified and extracted features and a k-Nearest Neighbors algorithm; ([Kumar ¶ 0062] as detailed above; “A machine learning algorithm may include supervised algorithms and/or unsupervised algorithms. Various types of algorithms may be used, such as…k-nearest neighbors” [Kumar ¶ 0092]).
However, the combination does not expressly teach validating the trained supervised learning model by using a separate data set to ensure a reliability in predictions by the supervised learning model.
In the same field of endeavor, LaViale teaches a means of implementing the k-nearest neighbors algorithm via a supervised learning model that validat[es] the trained supervised learning model by using a separate data set to ensure a reliability in predictions by the supervised learning model (“One common technique for choosing the optimal value of k is to use cross -validation, where we split the data into training and testing sets multiple times and evaluate the algorithm’s performance for different values of k” [LaViale page 8]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated validat[ing] the trained supervised learning model by using a separate data set to ensure a reliability in predictions by the supervised learning model as taught by LaViale into the combination because both Kumar and LaViale are directed towards implementing the k-nearest neighbors algorithm via a supervised learning model. As is commonly understood in the art, incorporating cross-validation as disclosed in LaViale would improve overall reliability / estimate accuracy of the supervised learning model (“Cross-validation provides a more reliable estimate of performance, but can be computationally expensive” [LaViale page 9]).
Regarding claim 3, the combination of Kumar, Tamayo, Chen, and LaViale teaches the limitations of parent claim 2, and Kumar further teaches continuously updating and refining the collected data and the supervised learning model based on new data about the areas of expertise, the past performances, and the client feedback to maintain an accuracy in a categorization of the partners based on the strengths of the partners (“A machine learning algorithm can be iterated to learn a target model f that best maps a set of input variables to an output variable, using a set of training data… The training data may be updated based on, for example, feedback on the accuracy of the current target model f. Updated training data may be fed back into the machine learning algorithms, which in turn updates the target model f.” [Kumar ¶ 0091]).
Regarding claims 9-10 and 16-17, they are apparatus and product claims that substantially correspond to the methods of claims 2-3, which are already taught by the combination of Kumar, Tamayo, Chen, and LaViale as detailed above. Consequently, they are rejected for the same reasons.
Claims 4-5, 11-12, and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar, Tamayo, and Chen, as applied to claims 1, 8, and 15 above, further in view of Li et al. (Pub. No. US 20170091193 A1, “Calculating an Influence Score of a Member”, published 03/30/2017), hereinafter Li.
Regarding claim 4, the combination of Kumar, Tamayo, and Chen teaches the limitations of parent claim 1, and Kumar further teaches collecting data about the interactions between the partners and the clients, the collected data including a frequency of the interactions, rates of success resulting from the interactions, and scores indicating feedback from the clients and the partners about the interactions (see opportunity parameters [Kumar ¶ 0030] and outcome indicators [Kumar ¶ 0051] as detailed in claim 1 above);
cleaning the collected data to remove outliers, manage missing values, and validate relationship metrics included in the collected data, wherein the generating the network graph is based on the cleaned data (see means of partitioning, incl. strict partitioning with outliers, in [Kumar ¶ 0064]);
Kumar further teaches edges in a graph being based on past interactions between the partners and the clients, the strengths of the partners, and measures of trustworthiness of the partners and the clients ([Kumar ¶ 0070] as detailed in claim 1 above).
However, the combination does not expressly teach estimating, using a graph database model, strengths of the connections between the partners and the clients by measuring weights of the edges in the network graph and using centrality measures, and generating, using the graph database model, a client-partner relationship index based on the estimated strengths of the connections between the partners and the clients.
In the same field of endeavor, Li teaches a means of evaluating relationships between entities that estimat[es], using a graph database model, strengths of the connections between entities by measuring weights of the edges in the network graph and using centrality measures, (“In some instances, the social graph data 214 can be based on an entity's presence within the online social network service. For example, consistent with some embodiments, a social graph is implemented with a specialized graph data structure in which various entities (e.g., people, companies, schools, government institutions, non-profits, and other organizations) are represented as nodes connected by edges, where the edges have different types representing the various associations and/or relationships between the different entities” [Li ¶ 0062]; “Moreover, the centrality score for the first member can be further dependent on a strength of connection value between the first member and a first-degree connection. As previously described, the strength of connection value can be dependent on a number of common connections between the first member and the first-degree connection. The contribution factor for each first-degree connection is weighted using the strength of connection value between the first member and the first-degree connection.” [Li ¶ 0082]) and
generat[es], using the graph database model, an entity-entity relationship index based on the estimated strengths of the connections between the partners and the clients (“Moreover, the influence score can be further based on a damping factor or a strength of connection value between the first member and a first-degree connection.” [Li ¶ 0078])
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated estimating, using a graph database model, strengths of the connections between the partners and the clients by measuring weights of the edges in the network graph and using centrality measures, and generating, using the graph database model, a client-partner relationship index based on the estimated strengths of the connections between the partners and the clients as taught by Li into the combination because both Kumar and Li are directed towards evaluating relationships between entities. Incorporating the teachings of Li would enable contextual evaluation of a partner’s performance scores based on their structural placement inside the communication pipeline ([Li ¶ 0014-0015]).
Regarding claim 5, the combination of Kumar, Tamayo, Chen, and Li teaches the limitations of parent claim 4, and Li further teaches continuously updating the entity-entity relationship index and the network graph by using data about new interactions between the partners and the clients to re-estimate the strengths of the connections between the partners and the clients (“Once the centrality score of each member in the network is determined, then the centrality score for each member is updated based on a change in the network. A change in the network occurs when a new member is added to the network, a member is deleted from the network, two members form a new connection, or two members remove a connection.” [Li ¶ 0037])
Regarding claims 11-12 and 18-19, they are apparatus and product claims that substantially correspond to the methods of claims 4-5 which are already taught by the combination of Kumar, Tamayo, Chen, and Li as detailed above. Consequently, they are rejected for the same reasons.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to VIJAY M BALAKRISHNAN whose telephone number is (571) 272-0455. The examiner can normally be reached 10am-5pm EST Mon-Thurs.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, JENNIFER WELCH can be reached on (571) 272-7212. 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.
/V.M.B./
Examiner, Art Unit 2143
/JENNIFER N WELCH/Supervisory Patent Examiner, Art Unit 2143
Prosecution Insights