System and method for evaluating compatibility between entities

Abstract

A computer-implemented method for evaluating compatibility between a first entity and a second entity, the method comprising: obtaining data associated with the first entity and data associated with the second entity; deriving a plurality of factors from the obtained data, wherein the plurality of factors comprise at least one factor representing an attribute of the first entity and at least one factor representing an attribute of the second entity; performing matrix operations that combine values of the plurality of factors and weights associated with the plurality of factors across dimensions and hierarchical levels to compute a matching score between the first entity and the second entity; and presenting the matching score and explainable insights indicating a contribution of each of the plurality of factors to the matching score.

Description

SYSTEM AND METHOD FOR EVALUATING COMPATIBILITY BETWEEN ENTITIESTechnical Field

[0001] This disclosure relates generally to computer-implemented matching systems and methods, and more specifically to computer-implemented matching systems and methods for evaluating compatibility between entities.Background

[0002] Computer-implemented systems are commonly used to evaluate compatibility between entities in applications such as recruitment, collaboration platforms and recommendation systems. These systems typically process data associated with different entities to determine potential matches.

[0003] Conventional matching systems often rely on relatively simple techniques, such as keyword comparisons, rule-based filtering or limited attribute scoring. These approaches evaluate attributes independently and may fail to capture complex relationships among multiple factors associated with the entities. Such systems may produce inaccurate or incomplete compatibility assessments, particularly where the evaluation involves multiple domains, hierarchical structures or contextual relationships between attributes. For example, in talent matching environments, conventional systems frequently prioritize static information such as qualifications or job titles while providing limited capability to analyse broader contextual characteristics, including behavioural attributes and long-term career trajectories.

[0004] There thus remains a need for improved computer-implemented systems and methods for evaluating compatibility between entities, capable of performing multidimensional and multi-hierarchical compatibility assessments.Summary

[0005] According to a first aspect, there is provided a computer-implemented method for evaluating compatibility between a first entity and a second entity, the method comprising: obtaining data associated with the first entity and data associated with the second entity; deriving a plurality of factors from the obtained data, wherein the plurality of factors comprise at least one factor representing an attribute of the first entity and at least one factor representing an attribute of the second entity;performing matrix operations that combine values of the plurality of factors and weights associated with the plurality of factors across dimensions and hierarchical levels to compute a matching score between the first entity and the second entity; andpresenting the matching score and explainable insights indicating a contribution of each of the plurality of factors to the matching score.

[0006] The method may further comprise dynamically adjusting the weights in response to changes in values of the plurality of factors associated with at least one of the first entity and the second entity.

[0007] The method may further comprise performing spectrum analysis on the plurality of factors to determine the weights for determining the matching score.

[0008] The method may further comprise processing the plurality of factors in a defined sequence based on an order of entry of the data, wherein the sequence is adjustable to influence the resulting matching score.

[0009] The method may further comprise including relationships among the plurality of factors based on hierarchical groupings of the factors in the matrix operations when determining the matching score.

[0010] The method may further comprise applying temporal weighting to the plurality of factors such that an influence of each factor on the matching score varies depending on at least one of a historical state, a current state and a predicted state of at least one of the first entity and the second entity.

[0011] The method may further comprise evaluating the plurality of factors in an acyclic manner.

[0012] The method may further comprise selecting different subsets of the plurality of factors for inclusion in computing the matching score.

[0013] The method may further comprise allowing the first entity and the second entity to modify data associated with the plurality of factors used in computing the matching score.

[0014] The method may further comprise:obtaining reference data from a plurality of reference sources associated with at least one of the first entity and the second entity;evaluating the reference data to generate a reference-based verification output; and incorporating the reference-based verification output into the matching score.

[0015] The explainable insights may comprise a breakdown of the plurality of factors used in the scoring, trend visualizations based on trend analysis outputs and reference-based verification results.

[0016] The method may further comprise:aggregating and analysing the plurality of factors, the trend analysis outputs and the reference-based verification results; andgenerating at least one of analytical reports and performance metrics derived from the matching score and the plurality of factors.

[0017] According to a second aspect, there is provided a computer- implemented matching system for evaluating compatibility between a first entity and a second entity, the system comprising:a data acquisition module configured to obtain data associated with the first entity and data associated with the second entity;a multi-factor scoring module configured to derive a plurality of factors from the obtained data, wherein the plurality of factors comprise at least one factor representing an attribute of the first entity and at least one factor representing an attribute of the second entityand to perform matrix operations that combine values of the plurality of factors and weights associated with the plurality of factors across multiple dimensions and hierarchical levels to compute a matching score between the first entity and the second entity; anda user interface module configured to present the matching score and explainable insights indicating a contribution of each of the plurality of factors to the matching score.

[0018] The multi-factor scoring module may be further configured to:dynamically adjust the weights in response to changes in values of the plurality of factors; andperform spectrum analysis on the plurality of factors to determine the weights for determining the matching score.

[0019] The multi-factor scoring module may be further configured to:process the plurality of factors in a defined sequence based on an order of entry of the data, wherein the sequence is adjustable to influence the resulting matching score;include relationships among the plurality of factors based on hierarchical groupings of the factors in the matrix operations when determining the matching score;apply temporal weighting to the plurality of factors such that an influence of each factor on the matching score varies depending on at least one of a historical state, a current state and a predicted state of at least one of the first entity and the second entity; andevaluate the plurality of factors in an acyclic manner.

[0020] The multi-factor scoring module may be further configured to:select different subsets of the plurality of factors for inclusion in computing the matching score; andallow the first entity and the second entity to modify data associated with the plurality of factors used in computing the matching score.

[0021] The system may further comprise a reference chain verification module configured to obtain reference data from a plurality of reference sources, generate a reference-based verification output and provide the reference-based verification output to the multi-factor scoring module for incorporation into the matching score.

[0022] The system may further comprise an analytics module configured to aggregate and analyse the plurality of factors, the trend analysis outputs and the reference-based verification results and generate at least one of analytical reports and performance metrics derived from the matching score and the plurality of factors.

[0023] According to a third aspect, there is provided a non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform the method in the first aspect.

[0024] According to a fourth aspect, there is provided an apparatus comprising one or more processors and a memory storing instructions that, when executed by the one or more processors, cause the apparatus to perform the method in the first aspect.

[0025] According to a fifth aspect, there is provided a computer program product, wherein the computer program product comprises instructions for performing the method in the first aspect.Brief Description of the Drawings

[0026] Exemplary embodiments of the present invention are hereinafter further described, by way of example only, with reference to the accompanying drawings, in which:

[0027] FIG. 1 is a schematic illustration of an exemplary system for evaluating compatibility between entities.

[0028] FIG. 2 is a flowchart of an exemplary method of evaluating compatibility between entities.Detailed Description

[0029] Referring to FIGS. 1 and 2, the present disclosure relates to a computer-implemented matching system 100 and its associated method (200) for evaluatingcompatibility between a first entity 10 and a second entity 20. The system 100 comprises a data acquisition module 106, a multi-factor scoring module 108 and a user interface module 110.

[0030] In general, the data acquisition module 106 obtains data associated with the first entity 10 and data associated with the second entity 20 (210). The multi-factor scoring module 108 computes the matching score based on the derived factors and associated weights (230). The user interface module 110 then presents the score along with detailed insights (240), allowing users to analyse and interpret the compatibility results effectively. Each component is configured to operate in an integrated manner to compute and present compatibility metrics. Together, these components provide a robust and flexible system for evaluating compatibility between entities 10, 20 with high accuracy, transparency and domain versatility, as will be described in greater detail below.

[0031] The system 100 and method (200) offer several advantages. First, by deriving a plurality of factors from entity data and performing multi-dimensional matrix operations, the system 100 and method 200 enable precise and nuanced evaluation of compatibility that reflects both individual attributes and interactions of the two entities 10, 20 10, 20 across hierarchical and dimensional contexts. Second, the system 100 and method (200) provide a flexible framework applicable to various domains, allowing evaluation of compatibility between individuals, organizations or other entities 10, 20. Third, the explainable output fosters trust and transparency, providing users with actionable insights rather than simple binary matches. Fourth, the modular architecture of the system 100 may support scalability and integration with external platforms, enabling dynamic updates to factors or weighting schemes while maintaining consistent computational rigor. East but not least, the modular separation between data acquisition and scoring logic also improves maintainability and allows the system 100 and method (200) to be readily adapted to different application domains. The proposed system 100 and method 200 thus improve scalability and allows the compatibility evaluation process to incorporate diverse information describing both entities 10, 20.Data Acquisition Module

[0032] The data acquisition module 106 is configured to obtain data associated with the first entity 10 and data associated with the second entity 20 (210). The data acquisition module106 may collect structured, semi-structured and / or unstructured data, including entity profiles, attributes, operational parameters, historical performance, functional characteristics, preferences and other relevant information. The data acquisition module 106 may collect information describing attributes of the entities 10, 20 from a variety of sources. Such sources may include user- submitted profiles, descriptions, uploaded records, online information sources, enterprise systems and external / internal databases, and so on.

[0033] In some embodiments, the data acquisition module 106 obtains data through market sector- specific question sets. These question sets are structured questionnaires tailored to industries or operational contexts, enabling the system 100 and method (200) to collect attributes that are relevant to the environment in which the compatibility evaluation is performed. The market sector-specific question sets may be stored as configurable templates and selected dynamically based on contextual parameters such as industry classification, operational domain or user selection. For example, a healthcare-related context may include questions relating to regulatory compliance, clinical practices and operational experience, whereas a technology-related context may emphasize technical capabilities, system design experience and familiarity with emerging technologies. Other sectors may similarly define questions relevant to their operational characteristics. The data acquisition module 106 may support optional domain- specific inputs for various industries or application contexts.

[0034] The data acquisition module 106 may further interact with data repositories through application programming interfaces (APIs). In one embodiment, the system 100 stores data within a database environment configured to store entity profiles, attributes and derived factors. Access to the database may be performed through a database access layer, which provides an abstraction interface between system modules and the underlying database technology. This abstraction layer allows the modules of the system 100 to retrieve and store information without knowledge of the internal structure or implementation of the database environment.

[0035] In particular, the database access layer may manage communication between the system modules and the database using APIs or other programmatic interfaces. Through these interfaces, the data acquisition module 106 can retrieve previously stored information, update entity data and store newly obtained information. The abstraction provided by the databaseaccess layer allows the system 100 to support multiple database technologies, including relational databases, distributed data stores and / or cloud-based storage environments.

[0036] The data acquisition module 106 may receive entity data from various input interfaces and / or external systems, process the incoming information and store the data within the database through the database access layer. In one embodiment as shown in FIG. 1, the data acquisition module 106 collects data directly from the first entity 10 and the second entity 20. The data acquisition module 106 may also collect the data from external sources, such as public databases, third-party databases, network-accessible repositories or historical records, to enrich the data set and improve the accuracy and robustness of the multi-factor scoring.

[0037] The data acquisition module 106 may also process the raw inputs to ensure completeness and normalization, preparing the data for analysis by the multi-factor scoring module 108. For example, the data acquisition module 106 may perform initial preprocessing operations, such as validating inputs, normalizing formats and organizing information into structured representations suitable for further analysis. The preprocessing operations may also include using natural language processing (NLP) techniques for extracting relevant information from unstructured text. Once stored, the data becomes accessible to the multifactor scoring module 108.Multi-factor Scoring Module

[0038] The multi-factor scoring module 108 derives a plurality of factors from the obtained data associated with the first entity 10 and the obtained data associated with the second entity 20, wherein the plurality of factors comprise at least one factor representing an attribute of the first entity 10 and at least one factor representing an attribute of the second entity 20 (220). The factors may be organized within a structured framework comprising a plurality of categories associated with each entity 10, 20.

[0039] In some embodiments, each of the plurality of factors derived from the obtained data is represented in a numerical form suitable for matrix computation. After the attributes associated with the first entity 10 and the second entity 20 are mapped to the corresponding factor categories, the multi-factor scoring module 108 assigns a quantitative value to eachfactor. The quantitative value may represent, for example, the presence, intensity, relevance or degree of correspondence of the factor with respect to each entity. These values may be normalized, scaled or transformed into numerical representations so that the plurality of factors can be organized into structured data vectors.

[0040] The numerical representations of the plurality of factors enable the multi-factor scoring module 108 to perform matrix operations that combine values of the plurality of factors and weights associated with the plurality of factors across dimensions and hierarchical levels to compute a matching score between the first entity 10 and the second entity 20. In this manner, attributes associated with the first entity 10 and the second entity 20 are converted from descriptive data into computational inputs suitable for mathematical processing. The matrix operations may therefore evaluate compatibility between the entities 10, 20 by computing relationships among the numerical factor values within a multi-dimensional matrix structure, thereby producing the matching score. These matrix operations provide a structured mechanism for integrating multiple attributes simultaneously, capturing complex interrelationships among factors and enabling quantitative evaluation of compatibility.

[0041] In some embodiments, the multi-factor scoring module 108 uses a structured MetaMatrix framework that enables compatibility assessment across multiple dimensions and hierarchical levels.

[0042] It will be appreciated that use of the Meta-Matrix framework is provided as an illustrative implementation. In other embodiments, the plurality of factors may be organized using alternative taxonomies, factor models or attribute classification structures suitable for representing relationships between the entities 10, 20. The factors may be arranged using other multi-dimensional frameworks, vector representations, graph-based structures or ontologybased classifications. Regardless of the particular representation used, the factors are converted into numerical values suitable for matrix operations that combine the factor values with associated weights across multiple dimensions and hierarchical levels to compute the matching score (230).

[0043] The Meta-Matrix representation provides a structured taxonomy that facilitates mapping between attributes of the entities 10, 20 and supports the matrix operations used tocompute the matching score. The advantages of using the proposed Meta-Matrix include improved accuracy, nuanced evaluation across multiple dimensions, transparency of the scoring process and the ability to dynamically accommodate evolving entity characteristics. Leveraging the structured factor framework, the multi-factor scoring module 108 generates high-precision, multi-dimensional matches that are adaptive, explainable and robust across diverse application domains.

[0044] The Meta-Matrix framework may be applied across various application contexts. Illustrative embodiments of the Meta-Matrix framework for different application contexts, i.e., human resource, organizational partnerships and personal compatibility, are described below.Meta-Matrix Framework: Example One - Human Resources (HR) Context

[0045] In an exemplary HR context, the Meta-Matrix is implemented as a multidimensional matrix (such as a 7x7, 5x5, 3x3, 77, 55, or 33matrix), where each element of the Meta-Matrix represents a combination of a factor from the first entity 10 (e.g., candidate) and a factor from the second entity 20 (e.g., employer), computed as the product of the factor values and an associated weight.. In a 7x7 matrix for example, candidate factors include “Character”, “Experience”, “Knowledge”, “Motivation”, “Skills”, “Talents” and “Values”, and employer factors include “Domain”, “Hierarchies”, “Functions”, “Dimensions”, “Roles”, “Structures” and “Systems”.

[0046] To be more specific, the candidate factor “Experience” may be evaluated against the employer factor “Hierarchies” to assess alignment between the candidate’s prior roles and the organizational level requirements. Similarly, the candidate factor “Skills” may be compared with the employer factor “Functions” to evaluate functional competency, and the candidate factor “Values” may be assessed against the employer factor “Dimensions” to determine compatibility in terms of long-term engagement potential, and so on, as shown in Table 1 below. Each pair of factors represents a conceptual relationship used by the multi-factor scoring module 108 to evaluate compatibility between the entities 10, 20.First entity Second entity(candidate) factors (employer) factorsCharacters RolesSkills FunctionsKnowledge DomainsExperience HierarchiesMotivation StructuresTalents SystemsValues DimensionsTable 1

[0047] The factor pair “Roles” and “Character” relates to alignment between organizational expectations and individual disposition. For the second entity 20 (in this example a potential employer), “Roles” define positions, responsibilities and organizational expectations associated with a particular function or responsibility within the organization. For the first entity 10 (in this example a candidate), the corresponding factor “Character” represents personal qualities, behavioural traits and ethical grounding that influence how an individual embodies and fulfils assigned roles. Evaluating these factors together allows the multi-factor scoring module 108 to determine whether the individual’s disposition and behavioural tendencies align with the expectations associated with the role.

[0048] The factor pair “Functions” and “Skills” relates to operational capability. For the employer 20, “Functions” represent operational task areas and responsibilities required to achieve organizational objectives. For the candidate 10, the corresponding factor “Skills” represents demonstrated proficiencies, technical capabilities and functional abilities enabling effective task execution. Although the concept of function may conceptually relate to skills, knowledge or experience, evaluating functions primarily in relation to “Skills” provides a practical and measurable basis for matching operational capability to job requirements.

[0049] The factor pair “Domain” and “Knowledge” relates to subject-matter expertise. For the employer 20, “Domain” represents the industry, sector or professional field in which the role operates. For the candidate 10, the corresponding factor “Knowledge” represents the intellectual foundation, educational background and applied familiarity with relevant subject areas. Evaluating domain primarily in relation to “Knowledge” allows the multi-factor scoring module 108 to assess subject-matter familiarity in a more precise and structured manner.

[0050] The factor pair “Hierarchies” and “Experience” relates to organizational level alignment. For the employer 20, “Hierarchies” represent organizational ranking, reporting structures, leadership levels and authority frameworks within which roles operate. For the candidate 10, the corresponding factor “Experience” represents the individual’s verified work history, progression of responsibilities and duration of service across prior roles. Evaluating these factors together allows the multi-factor scoring module 108 to determine whether the individual’s experience aligns with the hierarchical level associated with the opportunity.

[0051] The factor pair “Structures” and “Motivation” relates to compatibility between work environment and internal drive. For the employer 20, “Structures” represent organizational operating models, work arrangements and structural frameworks for organizing work, such as centralized or distributed teams, remote or on-site environments and agile or hierarchical management models. For the candidate 10, the corresponding factor “Motivation” represents internal drivers, psychological incentives and career ambitions influencing engagement with work. Evaluating these factors together enables the multi-factor scoring module 108 to determine whether the structural environment provided by the employer aligns with the motivational drivers of the candidate 10.

[0052] The factor pair “Systems” and “Talents” relates to the interaction between technological or operational systems and inherent individual capabilities. For the employer 20, “Systems” represent workflows, processes, tools and technological infrastructures used to achieve operational objectives. For the candidate 10, the corresponding factor “Talents” represents innate aptitudes, cognitive patterns and natural abilities that exist independently of learned skills, such as logical reasoning, pattern recognition, analytical thinking or creative problem- solving. Evaluating these factors together enables the multi-factor scoring module108 to assess whether the individual’s inherent aptitudes allow them to effectively operate within, optimize or design the systems used by the employer 20.

[0053] The factor pair “Dimensions” and “Values” relates to long-term alignment and broader organizational impact. For the employer 20, “Dimensions” represent broader evaluative aspects relating to organizational impact and development, including cultural alignment, adaptability, collaboration, cross-functional interaction and potential contribution beyond immediate role responsibilities. These dimensions may also reflect the depth and breadth of contributions expected over time. For the candidate 10, the corresponding factor “Values” represents personal beliefs, priorities and guiding principles influencing behaviour, decision-making, perseverance and long-term commitment. Evaluating these factors together allows the multi-factor scoring module 108 to determine whether the individual’s values align with the broader cultural and organizational dimensions associated with the employer 20.

[0054] By organizing factors according to these corresponding categories, the multi-factor scoring module 108 can perform matrix operations that systematically evaluate compatibility across multiple dimensions simultaneously. This structured factor framework enables the multi-factor scoring module 108 to incorporate technical competencies, organizational structure alignment, domain expertise, motivational drivers, cognitive aptitudes and cultural compatibility into a unified quantitative evaluation of compatibility between the candidate 10 and the employer 20.

[0055] In one embodiment, the Meta-Matrix is organized so that each row is associated with a candidate factor, and each column is associated with an employer factor, such that each element of the matrix represents the weighted combination of a candidate factor and an employer factor. For example, a subset of three candidate factors, i.e., “Experience”, “Skills” and “Values” and three employer factors, i.e., “Hierarchies”, “Functions” and “Dimensions” may form a 3x3 sub-matrix, where each element captures the contribution of a specific candidate-employer factor pair to the overall matching score. A simplified 3x3 subset of the Meta-Matrix can be represented mathematically as shown in matrix Ml below:^Exp ’ ^Hie ’W11 ^Exp ’ ^Fun ’VV12 ^Exp ’ ^Dim ’w13M = Q>ki ’ ^Hie ’W21 Q>ki ’ ^Fun ’w22 Q>ki ’ ^Dim ’w23 - (Ml)Cval ’ ^Hie ’W31 ^Val ’ ^Fun ’VV32 ^Val ’ ^Dim ’w33where CExp, CSkiand CVairepresent quantified values of the candidate’s factors “Experience”, “Skills” and “Values”, respectively, El. EVunand EDimrepresent quantified values of the employer factors “Hierarchies”, “Functions” and “Dimensions”, respectively and Wj_j represent the weight assigned to each candidate-employer factor pair. Each element of the matrix Ml captures the weighted contribution of a specific candidate factor relative to a specific employer factor, allowing a structured, multi-dimensional assessment of compatibility.

[0056] The overall matching score S between the candidate 10 and employer 20 may then be computed using equation El below:5 =Z- -(El)For example, if a candidate’s experience value CExp= 5, the normalized employer hierarchy value EHie= 0.8 and the weight assigned to this factor pair w1±= 0.6, the contribution to the overall matching score from this factor pair is 5 X 0.8 X 0.6 = 2.4 The multi-factor scoring module 108 repeats this computation across all factor pairs and sums the results to obtain the total matching score.

[0057] In some embodiments, the scoring algorithm within the Meta-Matrix framework incorporates additional features that enhance precision, adaptability and transparency of the compatibility evaluation. The weights associated with each factor pair can be dynamically adjusted in response to changes in candidate or employer data, enabling the system to reflect evolving profiles, preferences or organizational priorities over time. Spectrum analysis may be applied to determine the relative intensity or relevance of each factor, refining weight assignments during the matrix computation and ensuring that the resulting matching scores accurately reflect the multidimensional characteristics of each entity.

[0058] The multi-factor scoring module 108 may process the plurality of factors in a defined sequence based on the order in which data is entered, allowing the evaluation to account for narrative or temporal context in candidate profiles and employer requirements. Hierarchical relationships among the factors can also be incorporated into the matrix operations, providing a structured, multi-level assessment that respects dependencies and interactions across domains, functions and roles. Temporal weighting may further be applied, giving greater influence on historical performance or verified experience while also considering current competencies and future potential.

[0059] Subsets of factors can be selectively included in the computation to focus on specific aspects of compatibility, such as technical skills, leadership potential or cultural fit. The multi-factor scoring module 108 may allow candidates and employers to modify factor values, ensuring user-driven updates are reflected in real-time matching scores. Reference chain verification may be integrated, whereby reference data from multiple sources is evaluated for credibility and incorporated into the overall score, supporting trustworthiness and accountability.

[0060] Explainable insights may accompany the matching scores, providing a breakdown of contributions from each factor, trend visualizations derived from historical data and verification outputs from reference checks. Analytical reports and performance metrics may also be generated, offering detailed evaluations of candidate pipelines, hiring outcomes and organizational alignment. These features collectively enhance the accuracy, transparency and interpretability of the Meta-Matrix framework, enabling HR professionals to make informed, evidence-based decisions.Meta-Matrix Framework: Example Two -Commercial Partnership Context

[0061] In an exemplary commercial partnership context, the Meta-Matrix framework may be implemented as a multi-dimensional matrix (such as a 7x7, 5x5, 3x3, 77, 55, or 33matrix) and applied to evaluate compatibility between parties seeking commercial partnerships. In such cases, the first entity 10 may be a selling party (seller) offering products or services, and the second entity 20 may be a buying party (buyer) seeking to procure goods, services or collaborative capabilities.

[0062] Through this matrix-based evaluation, the multi-factor scoring module 108 can determine whether the capabilities and operational characteristics of the seller 10 align with the requirements and strategic objectives of the buyer 20.

[0063] The plurality of factors associated with the seller 10 may include attributes such as “Capability”, “Capacity” and “Reliability”. Correspondingly, the plurality of factors associated with the second entity 20 (i.e., buyer) may include attributes such as “Requirement”, “Demand” and “Strategy”. These factors may be derived from data obtained from the respective entities 10, 20, including supplier profiles, procurement specifications, operational requirements and partnership objectives.

[0064] For example, the factor pair “Capability” and “Requirement” represents the alignment between the technical and operational competencies of the seller 10 and the specific functional requirements of the buyer 20. The system assesses whether the seller possesses the necessary capabilities to satisfy the buyer’s technical and operational expectations.

[0065] The factor pair “Capacity” and “Demand” evaluates whether the seller’s operational scale, production throughput or service bandwidth is sufficient to meet the buyer’s expected procurement volume or service utilization.

[0066] Similarly, the factor pair “Reliability” and “Strategy” measures whether the historical performance, consistency and dependability of the seller align with the buyer’s longterm procurement objectives and strategic goals.

[0067] A simplified subset of the Meta-Matrix framework may be represented as a 3x3 matrix, where each row corresponds to a factor associated with the seller 10 and each column corresponds to a factor associated with the buyer 20. For example, seller factors “Capability”, “Capacity” and “Reliability” may correspond to the rows of the matrix, while buyer factors “Requirement”, “Demand” and “Strategy” may correspond to the columns. The matrix representation may be expressed as shown in matrix M2 below:C '-cap ’ Breq • W- Ccap ’ Bdem ’ ^12 Ccap ’ Bstr• W13M = r'-cpt ’ Breq • W21Ccpt ’ Bdem ’ W22Ccpt ’ Bstr• W23—(M2) Orel ’ Breq • W31Crel ’ Bdem ’32Crel ’ Bstr• W33where Ccap, Ccptand Crelrepresent quantified values of the seller’s factors “Capability”, “Capacity” and “Reliability”, respectively and Breq, Bdemand Bstrrepresent quantified values of the buyer’s factors “Requirement”, “Demand” and “Strategy”, respectively. The termrepresents the weight assigned to each seller-buyer factor pair.

[0068] The overall compatibility score S may then be computed using equation E2 below:5=Z- — (E2)For example, if the seller’s “Capability” value is 7, the normalized buyer “Requirement” value is 0.9 and the weight assigned to this factor pair w1±= 0.5, the contribution to the overall matching score from this factor pair is 7 X 0.9 X 0.5 = 3.15. The multi-factor scoring module 108 repeats this computation across all factor pairs and sums the results to obtain the total matching score between the seller 10 and the buyer 20.

[0069] The multi-factor scoring module 108 may allow dynamic adjustment of factor weights based on real-time changes in seller or buyer attributes, ensuring that the partnership score reflects current conditions and priorities. Spectrum analysis may be applied to evaluate the relative intensity or relevance of each factor pair, refining the contribution of each sellerbuyer interaction to the overall score. The multi-factor scoring module 108 can selectively process subsets of factors most relevant to a particular project or negotiation, enabling focused evaluation that considers only critical attributes. Temporal weighting may also be applied to account for historical performance, current capabilities and projected capacities, giving higher influence on verified past reliability while still considering anticipated performance and strategic alignment.

[0070] Data secrecy and confidentiality are maintained throughout the evaluation process. The multi-factor scoring module 108 may implement differential privacy techniques,introducing controlled randomness to computed scores to prevent disclosure of sensitive underlying data. Alternatively, federated computation approaches may be used, allowing score calculations to occur without centralized access to raw seller or buyer factor values. All factor data, matrices and scores are transmitted and stored using secure encryption protocols, ensuring that sensitive business information remains protected from unauthorized access. Reference data obtained from external sources is also handled securely and verification processes ensure the integrity and credibility of such data before it influences the partnership score.Meta-Matrix Framework: Example Three - Personal Compatibility Context

[0071] In an exemplary personal compatibility context, the Meta-Matrix framework is applied to evaluate compatibility between a first individual 10 and a second individual 20, for example, for use in matchmaking or other personal compatibility platforms. The plurality of factors associated with the first individual 10 may include attributes such as “Character”, “Motivation” and “Values”, reflecting personality traits, internal drivers and personal priorities. Correspondingly, the plurality of factors associated with the second individual 20 may include “Character”, “Motivation” and “Values”, representing the traits, drivers and priorities of the counterpart.

[0072] This framework allows assessment of compatibility based on multiple personality and motivational dimensions, providing explainable insights into how individual traits align across a structured, multi-dimensional matrix. Temporal weighting and spectrum analysis may further refine the matching score by emphasizing more stable traits over transient preferences.

[0073] The Meta-Matrix is now implemented as a 3x3 matrix for illustration, where each row is associated with a factor of the first individual 10 and each column is associated with a factor of the second individual 20, such that each matrix element represents the weighted combination of a specific factor pair. For example, the matrix representation may be expressed as shown in matrix M3 below:~Q'ha ’ 4'ha ’W11 Q'ha ’ 41ot ’VV12 Q'ha ’ ^Val ’w13"M = Gvlot ’ 4'ha ’VV21 Gvlot ’ 41ot ’w22 Gvlot ’ ^Val ’VV23 -(M3)-Cval ’ 4'ha ’W31 ^Val ’ 41ot ’w32 Cval ’ ^Val ’w33 -where Ccha, CMotand CVairePresent quantified values of the first individual’s “Character”, “Motivation” and “Values”, respectively, while factors, / Cha, / Motand / Vairepresent quantified values of the second individual’s “Character”, “Motivation” and “Values”, respectively and W(7- represents the weight assigned to each factor pair.

[0074] The overall matching score S is computed as the sum of all weighted factor pair products using equation E3 below:S = — (E3)For example, if the first individual’s “Character” value is 4, the second individual’s “Character” value is 0.9 and the weight w1±= 0.7, the contribution to the overall matching score from this factor pair is 4 X 0.9 X 0.7 = 2.52. The system 100 repeats this calculation for all factor pairs and sums the results to obtain the total compatibility score.

[0075] The multi-factor scoring module 108 preferably allows each individual 10, 20 to update or refine their factor values, ensuring that evolving preferences, experiences or traits are reflected in the compatibility score. The weights associated with each factor pair may be dynamically adjusted based on changes in the factor values, spectrum analysis of trait intensity or relevance to the overall compatibility evaluation. Temporal weighting may be applied so that past verified experiences or demonstrated traits are given higher influence than current preferences or future aspirations, allowing the compatibility score to reflect both proven capabilities and developmental trajectory. The multi-factor scoring module 108 may evaluate factors in an acyclic manner to prevent circular dependencies between factor relationships and can select different subsets of factors for inclusion in the calculation depending on context, relevance or user-defined priorities.

[0076] Reference data may be obtained from multiple sources associated with one or both individuals 10, 20 to generate a reference-based verification output, which can be incorporated into the compatibility score to enhance reliability and trustworthiness. Personal factor values, weights, temporal information and reference verification results, may be processed and storedwith robust data secrecy measures, such as encryption, differential privacy and federated learning, ensuring that sensitive personal information remains protected while still enabling accurate and explainable compatibility evaluation. Explainable insights are preferably generated alongside the compatibility score, including visualizations of factor pair contributions, temporal influences, spectrum-based analyses and reference verification results, providing transparency and interpretability of the matching process without exposing confidential information.Dynamic Weight Adjustment Mechanisms

[0077] The multi-factor scoring module 108 is preferably further configured to dynamically adjust the weights in response to changes in values of the plurality of factors associated with at least one of the first entity 10 and the second entity 20.

[0078] The multi-factor scoring module 108 may assign higher weight values to factors representing verified, historical performance (such as prior roles, completed projects and documented competencies) and lower weight values to factors representing future preferences that have not yet been realized. In an exemplary HR matching embodiment, factors corresponding to the experience of a candidate are weighted three points per match, while factors corresponding to preferred future fields are weighted one point per match. This explicit weighting ensures that the matching score prioritizes proven capability over aspirational interest while still accounting for an individual’s intended career trajectory.

[0079] The multi-factor scoring module 108 may incorporate adaptive learning mechanisms, wherein the relative weights of factors are continuously updated based on newly obtained data, historical trends and inferred patterns. Adjustable weighting may be applied to account for historical states, current states and predicted future states of both entities 10, 20, such that the influence of each factor on the computed matching score varies. For instance, in an exemplary HR context, a candidate who has accumulated significant experience in a particular role or domain may receive greater influence in the score, whereas indicated preferences for future roles are credited proportionally less. This adaptive approach allows reflecting the evolving profiles while maintaining robustness against overemphasis on unverified inputs.

[0080] In implementing the disclosed dynamic weight adjustment in the HR context, the multi-factor scoring module 108 may preferably also incorporate career trajectory considerations. In such embodiments, the proposed algorithm of the multi-factor scoring module 108 evaluates sequences of past roles, durations, transitions across domains and progression in hierarchical levels to predict the likely future direction of a candidate’s career. For example, if a candidate 10 initially trained in information technology (IT) but subsequently worked in the banking sector, the system 100 analyses this progression to determine whether the candidate 10 seeks IT roles within the financial sector or is pursuing a broader career transition, and can adjust weights to favour matches in IT within financial services while still accounting for broader career transitions. Such trajectory-based adjustments improve the relevance and accuracy of the computed matching score while enabling forward-looking predictions of compatibility.Spectrum Analysis Mechanisms

[0081] The multi-factor scoring module 108 is preferably further configured to perform spectrum analysis on the plurality of factors to determine weights for calculating the matching score between the first entity 10 and the second entity 20. Spectrum analysis provides a continuous evaluation of traits, preferences and capabilities along defined dimensions rather than using binary or discrete categorizations. Each factor associated with an entity is represented along a spectrum, capturing gradations and subtleties in attributes that influence compatibility. By mapping traits to continuous scales, the multi-factor scoring module 108 can capture nuanced differences between the first entity 10 and the second entity 20 that conventional categorical matching fails to reflect.

[0082] In an exemplary HR context, spectrum-based factors may include one or more of task orientation, social interaction preference, decision-making approach, work pace preference and environmental adaptability. Task orientation evaluates a candidate’s propensity toward precision versus flexibility in executing responsibilities. Social interaction preference reflects tendencies for collaborative engagement versus independent work. Decision-making approach quantifies the degree to which a candidate 10 relies on analytical reasoning versus intuitive judgment. Work pace preference measures steadiness versus variability in handling workload overtime and environmental adaptability assesses an individual’s capacity to operateeffectively in structured versus dynamic work settings. These factors are quantified along continuous scales, allowing the multi-factor scoring module 108 to capture subtle differences in performance style and compatibility with organizational expectations.

[0083] The spectrum analysis is preferably integrated with matrix operations, wherein values derived from spectrum-based factors may be weighted across hierarchical levels and dimensions. The resulting weighted values contribute to the overall matching score, reflecting both the intensity and alignment of traits. Furthermore, the multi-factor scoring module 108 preferably adjusts factor weights dynamically, considering temporal relevance, career trajectory considerations and evolving data associated with the entities 10, 20. This ensures that the computed score is sensitive to context, while maintaining a robust representation of the characteristics of the entities 10, 20.

[0084] In an HR matching example, spectrum analysis mechanisms include a multidimensional evaluation of leadership skills, incorporating multiple dimensions and hierarchical sub-factors. Leadership may be assessed across various dimensions, such as inspiration and decision-making, each represented on a normalized scale (e.g., between 0 and 1). Within each dimension, sub-factors are organized hierarchically, vision communication and motivation techniques, each having an assigned score (e.g., between 0 and 1) representing the entity’s performance. The leadership factor matrix L may be represented as shown in matrix M4 below:0.8 0.7 0.9L = 0.6 0.8 0.7 — (M4).0.9 0.7 0.8.where each row corresponds to a leadership dimension (e.g., inspiration, decision-making, strategic thinking) and each column corresponds to a hierarchical sub-factor (e.g., vision communication, motivation techniques and execution ability). The relative importance of each dimension is captured in a weight matrix Wd= [0.4, 0.3, 0.3] and the importance of each subfactor within a dimension is represented in a sub-factor weight matrix as shown in matrix M5 below:[0.4 0.3 0.30.4 0.3 0.3 — (M5)0.3 0.3.

[0085] The multi-factor scoring module 108 preferably computes the final leadership score by performing matrix multiplication of the leadership factor matrix L with the sub-factor weights Wsand dimension weights Wd, yielding a final leadership score of 0.779, reflecting the candidate’s capabilities across multiple leadership dimensions and hierarchical sub-factors.

[0086] To ensure explainability and transparency, the results of spectrum analysis may be presented to the user through the user interface module 110. Visualizations depict the continuum of each factor, highlighting where the first entity 10 and the second entity 20 lie along each spectrum. The user interface module 110 may also provide textual insights describing how the factor contributes to the overall matching score.

[0087] In some embodiments, the spectrum analysis further comprises identifying latent factors underlying the plurality of factors and performing dimensionality reduction to improve the quality of the data used in scoring. For example, techniques such as eigenvalue decomposition, singular value decomposition, or similar matrix factorisation methods may be applied to extract underlying patterns and correlations among factors. By projecting the data into a reduced-dimensional space, redundant or less informative components may be removed, thereby advantageously minimizing noise and enhancing the robustness and stability of the matching score. This process enables the system 100 to capture meaningful relationships that may not be explicitly represented in the original data while improving computational efficiency.

[0088] The use of spectrum analysis mechanisms provides several advantages over conventional binary or threshold-based matching systems. In the HR context, by capturing variations along continuous dimensions, the system 100 can recognize candidates who may not meet exact predefined criteria but possess complementary traits that indicate potential for high compatibility. Spectrum analysis, in combination with multi-factor scoring and temporal weighting (which will be discussed later), advantageously allows for personalized and context-aware matches that account for both current abilities and aspirational potential.Data Entry Sequence Processing Mechanisms

[0089] The multi-factor scoring module 108 is preferably further configured to process the plurality of factors in a defined sequence based on an order of entry of the data, wherein the sequence is adjustable to influence the resulting matching score. The data acquisition module 106 may record the sequence in which these data are entered or received. This order of entry forms part of the contextual information used by the multi-factor scoring module 108 during evaluation.

[0090] The multi-factor scoring module 108 may utilize the recorded order of entry when performing matrix operations that combine factor values and associated weights. By processing factors according to the defined sequence, the multi-factor scoring module 108 is able to interpret the relative importance of factors within the broader context of each entity. Factors entered earlier may influence the prioritization or interpretation of factors entered later, thereby enabling the multi-factor scoring module 108 to construct a contextual narrative of the entities 10, 20’ attributes.

[0091] This sequence-adjustable processing mechanism supports narrative formation, allowing the multi-factor scoring module 108 to analyse in a manner that reflects the temporal and contextual relationships among the factors. The multi-factor scoring module 108 may apply matrix operations that consider not only the values of each factor but also the sequencedependent context, which may affect hierarchical groupings, temporal weighting and spectrum analysis applied in later stages. The multi-factor scoring module 108 thereby ensures that the computed matching score represents a context-sensitive and nuanced assessment of compatibility between the first entity 10 and the second entity 20.

[0092] In some HR matching applications, the defined sequence of data entry allows the multi-factor scoring module 108 to distinguish between factors representing demonstrated past experience and factors representing future-oriented preferences. The multi-factor scoring module 108 may assign greater influence on factors corresponding to verified past experience, while factors representing aspirational or preferred future attributes are incorporated with comparatively lower influence. This approach ensures that the matching score primarilyreflects proven capabilities while still allowing the multi-factor scoring module 108 to consider potential career direction or progression.

[0093] The same sequence-based processing principle may be applied when evaluating other types of entities 10, 20, such as sellers and buyers as well as two individuals. In such cases, the order in which factor data is entered enables the multi-factor scoring module 108 to prioritize certain categories of attributes during evaluation.

[0094] By carefully orchestrating the sequence of trait analysis, the multi-factor scoring module 108 can improve both the efficiency of the matching process and the quality and relevance of the matches produced. The multi-factor scoring module 108 preferably processes the plurality of factors according to the defined sequence based on the order of entry of the data. Through this sequence-based processing, the multi-factor scoring module 108 can interpret relationships among the factors and prioritize evaluation steps in a structured and context-aware manner. Compared with conventional matching algorithms that process attributes without considering their order of entry, this sequencing approach enables more meaningful interpretation of factor relevance and improves the resulting matching score.

[0095] In one embodiment where trait matching algorithms is applied, the multi-factor scoring module 108 may apply a plurality of matchings to compare the factors derived from the data associated with the first entity 10 and the second entity 20. These operations may include word- weight matching, general matching, specific matching and range matching. Word-weight matching evaluates the semantic correspondence of terms and assigns weights according to contextual relevance. General matching evaluates broader factor categories to determine overall alignment between the entities 10, 20. Specific matching evaluates precise attributes associated with factor dimensions. Range matching evaluates compatibility when factor values fall within defined numerical ranges or spectrum intervals.

[0096] By integrating the defined sequence of factor processing with these matching operations, the multi-factor scoring module 108 enhances computational efficiency and improves interpretability of the resulting matching score. The sequencing strategy enables multi-factor scoring module 108 to evaluate factors progressively and contextually, therebyproducing more accurate and relevant compatibility evaluations between the first entity 10 and the second entity 20.Hierarchical Factor Relationship Processing Mechanisms

[0097] In some embodiments, the multi-factor scoring module 108 may be configured to include relationships among the plurality of factors based on hierarchical groupings of the factors when determining the matching score. Rather than evaluating each factor independently, the multi-factor scoring module 108 may organize related factors into hierarchical structures so that contextual relationships between attributes of the first entity 10 and the second entity 20 can be considered during compatibility evaluation.

[0098] The hierarchical structure may be implemented through the matrix operations, which categorizes attributes into structured layers such as domains, hierarchies and functions. Factors such as domains, hierarchies and functions form structured relationships that enable hierarchical evaluation during matrix operations. The domain generally represents a broad field or area of activity, a hierarchy represents a level of responsibility or authority within that field, and a function represents specific operational tasks or capabilities. By organizing factors in this manner, the multi-factor scoring module 108 can evaluate compatibility at multiple levels simultaneously.

[0099] During matrix operations, the multi-factor scoring module 108 respects the hierarchical groupings by applying weighting and relational rules that reflect the position of each factor within the hierarchy. For instance, factors at higher levels of the hierarchy, such as “Leadership”, may influence the scoring of subordinate functions, such as “Technical Implementation”. Similarly, related functions within the same hierarchy can interact to produce a composite score for that hierarchy, which then contributes to the overall matching score.

[0100] Consider an employer 10 specifying requirements for the domain “Engineering”, hierarchy “Strategic” and function “Systems Design”. The multi-factor scoring module 108 retrieves corresponding factors of a candidate 20, e.g., domain knowledge in engineering, experience at a strategic level and demonstrated systems design skills. Each factor is multiplied by a weight corresponding to its hierarchical level and then combined via matrix operations.The multi-factor scoring module 108 may also consider interactions across hierarchical levels. For example, a candidate’s experience in operational-level engineering projects may partially contribute to the strategic-level scoring. The resulting matching score reflects both the candidate’s direct alignment with the hierarchical requirements and their broader contextual suitability based on hierarchical interdependencies.

[0101] By incorporating hierarchical factor relationships in matrix operations, the multifactor scoring module 108 achieves improved accuracy and interpretability in matching outcomes. Hierarchical groupings ensure that the multi-factor scoring module 108 accounts for both explicit and implicit relationships among factors, enabling nuanced scoring across complex structures. In HR matching embodiments, hierarchical relationships ensure that organizational structures, leadership levels and functional responsibilities are properly aligned with candidate profiles.Temporal Weighting Mechanisms

[0102] The multi-factor scoring module 108 may be configured to apply temporal weighting to the plurality of factors, such that the influence of each factor on the matching score varies depending on at least one of a historical state, a current state or a predicted future state of the first entity 10 and the second entity 20. Temporal weighting allows the multi-factor scoring module 108 to incorporate not only the present characteristics of the entities 10, 20 but also their demonstrated historical performance and anticipated future trajectory, thereby enhancing predictive accuracy of the matching outcomes.For example, in an exemplary HR context, when evaluating a candidate’s aircraft engineering experience, the multi-factor scoring module 108 may consider three temporal phases: “Past” (for experience more than 5 years ago), “Recent” (for experience between 2 and 5 years ago) and “Current” (for experience between 0 and 2 years ago). For each phase, the multi-factor scoring module 108 evaluates key competencies, such as design skills, quality assurance and regulatory compliance.

[0103] The above competencies may be represented in a multi-temporal matrix E. where rows correspond to temporal phases and columns correspond to competencies, as shown in matrix M6 below:0.6 0.7 0.5 0.40.7 0.8 0.7 0.60.9 0.8 0.8 0.7.

[0104] The multi-factor scoring module 108 applies weights wtemporalto the temporal phases as [0.2, 0.3, 0.5] reflecting the increasing significance of more recent experience. Weights wcompetencyfor individual competencies are applied as [0.3, 0.3, 0.2, 0.2], reflecting the relative importance of each skill for the role. Using matrix multiplication, a final experience score is computed using equation E4 below:Final Score^temporal ’ E ’ ^competency (E4)The calculation yields a final experience score of 0.735, capturing both the candidate’s growth over time and current proficiency in aircraft engineering.

[0105] The proposed mechanism allows the multi-factor scoring module 108 to differentiate between past capability, recent development and current readiness, providing a more accurate and predictive evaluation than static or single-point scoring methods.

[0106] Consider another HR example where an employer seeking expertise in cloud infrastructure within the IT sector. A candidate has prior experience in on-premises infrastructure but has recently undertaken cloud training. The multi-factor scoring module 108 analyses historical experience, current capabilities and predicted future demand in the IT sector. Temporal weighting assigns the highest influence on verified historical experience, a moderate influence on current certifications and a forward-looking influence on predicted demand trends. The resulting matching score balances proven competence with adaptability to emerging industry requirements, enabling the multi-factor scoring module 108 to identify candidates who are not only currently capable but also positioned to meet future organizational needs.

[0107] By applying temporal weighting, the multi-factor scoring module 108 enhances the robustness and foresight of the matching process. It allows decision-makers to prioritize candidates or entities 10, 20 based on a comprehensive view that integrates verified past performance, current suitability and anticipated future alignment. Temporal weighting also operates in conjunction with hierarchical factor relationships, spectrum analysis and reference chain verification to produce an explainable compatibility score. This approach ensures that the matching system 100 is adaptable across various use cases.Acyclic Evaluation Processing Mechanisms

[0108] In some embodiments, the multi-factor scoring module 108 may be further configured to evaluate the plurality of factors in an acyclic manner, ensuring that factor dependencies do not create feedback loops or circular influences that could distort the computed matching score. This acyclic evaluation enables the multi-factor scoring module 108 to process complex interrelationships among factors while maintaining computational integrity and interpretability.

[0109] By evaluating the plurality of factors in an acyclic manner, the multi-factor scoring module 108 can model the factors as a factor graph, where each node represents a distinct factor derived from entity data and edges represent functional or hierarchical dependencies between factors. By enforcing acyclicity in this graph structure, the proposed algorithm can prevent recursive weighting or double counting of related factors, ensuring that each factor’s contribution to the matching score is independent and clearly interpretable. This approach is especially important in multi-dimensional applications where entities 10, 20 may have overlapping attributes across multiple domains, hierarchies or functional roles.

[0110] Consider an HR matching scenario where a candidate’s skills, experience and motivation are being evaluated against an employer’s role, hierarchy and system requirements. The candidate’s skills may influence their experience weighting and experience may in turn influence the evaluation of motivation. Without acyclic evaluation, these interdependencies could result in recursive amplification of certain factors. By constructing the factors as acyclic, it is ensured that each factor is evaluated once in the correct sequence, preserving the accuracy.

[0111] By implementing acyclic factor evaluation, the multi-factor scoring module 108 achieves robust and reproducible matching outcomes. This capability integrates seamlessly with hierarchical factor relationships, temporal weighting, spectrum analysis and reference chain verification, forming a comprehensive and transparent framework for multi-factor, multidimensional entity compatibility assessment. The acyclic evaluation feature is also applicable in other use cases such as vendor selection and any other multi-factor matching domain requiring precise factor independence and integrity in scoring.Factor Subset Selection Mechanisms

[0112] In some embodiments, the multi-factor scoring module 108 may be configured to select different subsets of the plurality of factors for inclusion in computing the matching score. The multi-factor scoring module 108 achieves this by determining which factors derived from the obtained data are relevant for the specific context in which compatibility between the first entity 10 and the second entity 20 is being evaluated.

[0113] The data acquisition module 106 may obtain information through structured questionnaires or data inputs tailored to operational domains. Based on the domain context, the multi-factor scoring module 108 may select a subset of factors to be included in the matrix operations used to compute the matching score. Factors that are not relevant to the selected domain may be excluded from the computation.

[0114] In one embodiment involving a medical domain, the multi-factor scoring module 108 may select factors relating to medical specialization, clinical experience and regulatory compliance. In one embodiment involving an IT domain, the selected subset of factors may instead include programming skills, system architecture knowledge and cybersecurity expertise. In one embodiment involving an education domain, the multi-factor scoring module 108 may select factors associated with teaching methodology, subject matter expertise and institutional hierarchy.

[0115] By selecting only relevant subsets of the plurality of factors, the multi-factor scoring module 108 improves computational efficiency and ensures that the resulting matching score reflects the attributes most pertinent to the specific matching scenario. This selectiveevaluation also enables the multi-factor scoring module 108 to operate across multiple domains while maintaining a consistent multi-factor scoring framework.Data Modification Mechanisms

[0116] In some embodiments, the data acquisition module 106, the multi-factor scoring module 108 and / or the user interface module 110 may be configured to allow the first entity 10 and the second entity 20 to modify data associated with the plurality of factors used in computing the matching score.

[0117] The ability to modify factor data may support ongoing profile updates, allowing entities 10, 20 to reflect changes in circumstances, capabilities or objectives over time. In an HR matching application, a candidate 10 may update a profile after acquiring new qualifications, completing a project or gaining additional experience within a particular domain. Similarly, an employer 20 or organizational entity 20 may revise role descriptions, functional requirements or organizational attributes as business needs evolve. The system 100 records such modifications and processes the updated data through the data preprocessing module before it is incorporated into the matrix operations of the multi-factor scoring module 108.

[0118] The system 100 may further allow entities 10, 20 to specify career aspiration fields or other forward-looking attributes. These fields may represent desired roles, preferred domains, aspirational hierarchies or future-oriented objectives that have not yet been realized but may influence the compatibility evaluation. Such fields enable the system to capture potential career trajectories or anticipated organizational developments while maintaining a distinction between aspirational attributes and demonstrated capabilities.

[0119] By allowing both entities 10, 20 to modify the data associated with the plurality of factors, the system 100 provides a flexible and adaptive framework for compatibility evaluation. This ensures that the matching score reflects the most current attributes of the entities 10, 20 while preserving transparency and user control over the information used in the matching process.Trend Analysis Module

[0120] In some embodiments, the system 100 may further comprise a trend analysis module (not shown in FIG. 1) configured to analyse patterns and developments associated with the plurality of factors over time. The trend analysis module identifies and incorporates relevant market trends, industry dynamics and external contextual factors in order to enhance evaluation of compatibility between entities 10, 20.

[0121] The trend analysis module may introduce additional dimensions and measurements into the matching process. These dimensions may include temporal dimensions, cross-domain relationships, and structural relationships associated with different roles and industries. The analysis may consider information across multiple sectors to identify transferable expertise and emerging opportunities across domains. The result of the trend analysis may further be used to adjust an initial matching score calculated by the multi-factor scoring module 108.

[0122] In some embodiments, the trend analysis module may analyse relationships among the factors using set operations such as intersections, unions, and differences to identify compatibility patterns. The trend analysis module may further infer attributes that are not explicitly specified in the entity data based on correlations within available data or external information sources. Outputs generated by the trend analysis module may be used to generate trend visualizations and explanatory insights associated with the matching score.Reference Chain Verification Module

[0123] In some embodiments, the system 100 may further comprise a reference chain verification module (not shown in FIG. 1) configured to obtain reference data from a plurality of reference sources associated with at least one of the first entity 10 and the second entity 20, and to generate a reference-based verification output and provide the reference-based verification output to the multi-factor scoring module 108 for incorporation into the matching score. The reference chain verification module may collect reference information from individuals or other entities who have prior knowledge of the first entity 10 and / or the second entity 20, such as supervisors, collaborators, clients or other professional contacts.

[0124] The system 100 may evaluate the credibility of each referee within the reference chain and assigns a dynamic credibility score based on factors such as historical reliability, hierarchical position, and interconnections with other references. The reference-based verification output is then generated by aggregating the reference data, with the mathematical strength of each contribution weighted according to the calculated credibility scores. This approach ensures that verification results reflect both the quality and trustworthiness of the sources, thereby enhancing the reliability of the overall matching score.

[0125] The reference chain verification module can evaluate the obtained reference data using a plurality of reference evaluation metrics designed to assess behavioural reliability and performance history. In an HR matching application, these metrics may include integrity, credibility, competence, staying power, restlessness, diversity and cohesiveness of experience.

[0126] Integrity may reflect the consistency and honesty of the entity’s representations. Credibility may measure the reliability of claims regarding qualifications or experience. Competence may evaluate the entity’s ability to perform functions associated with the relevant domain. Additional metrics may analyse patterns of professional engagement. For example, staying power may indicate stability and duration of participation in roles or projects, while restlessness may reflect frequent changes in positions or domains. Diversity of experience may represent the breadth of exposure across domains or functions, whereas cohesiveness of experience may evaluate whether the entity’s historical trajectory demonstrates logical progression or strategic development.

[0127] The reference chain verification module aggregates the above metrics to generate the reference-based verification output, which may include a composite verification score and associated explanatory information. This output is then provided to the multi-factor scoring module 108, which incorporates the verification results into the matrix operations used to compute the matching score.

[0128] In one embodiment, the reference chain verification module applies a verification matrix representing credibility scores for references associated with different credential types. For example, the verification matrix may be represented as shown in matrix M7 below:0.9 0.8 0.7 0 00.8 0.9 0.8 0.7 0 — (M7)0.7 0.8 0.9 0.8 0.7In this matrix, each row may correspond to a different credential type, such as academic degrees, professional certifications or specialized training, while each column corresponds to a reference position in the verification chain.

[0129] The reference chain verification module may further apply credential weights and reference position weights to reflect the relative importance of different credential categories and the reliability of different reference sources. In one embodiment, the credential weights are represented as: Wc= [0.3 0.3 0.4] and the reference position weights are represented as: Wr= [0.3 0.250.2 0.150.1]. The reference chain verification module performs matrix operations, such as matrix multiplication, to combine the verification matrix with the credential weights and reference weights. Through these operations, the reference chain verification module aggregates credibility scores across both credential categories and reference sources to generate a composite verification score.

[0130] In the given example, the calculation yields a final verification score of 0.714, indicating a relatively high level of confidence in the accuracy of the entity’s claimed credentials. This verification score may then be provided to the multi-factor scoring module 108, where it is incorporated into the overall compatibility evaluation between the first entity 10 and the second entity 20.

[0131] In some embodiments, the reference-based verification output is generated using secure communication interfaces and distributed ledger technologies. For example, the system 100 may utilise APIs to obtain reference data from external or third-party sources in a secure and authenticated manner. The obtained reference data may be recorded, validated, or crosschecked using cryptographic ledger mechanisms, such as blockchain-based records, to ensure immutability and traceability of the verification process. By leveraging cryptographic hashing, consensus protocols, or similar integrity-preserving techniques, the system 100 can prevent unauthorised modification of reference data and reduce the risk of fraudulent inputs, thereby enhancing the reliability and trustworthiness of the reference-based verification output.

[0132] By incorporating reference chain verification, the system 100 can supplement selfreported data with independently obtained information, thereby improving the reliability, transparency and predictive value of the compatibility evaluation.User Interface Module

[0133] The user interface module 110 presents the resulting matching score and provides explainable insights detailing how each factor contributes to the score (240). Explainable insights may include visualizations, graphical representations, numerical breakdowns or textual descriptions, allowing users to interpret and understand the rationale behind the computed compatibility. This transparency addresses limitations of traditional matching systems that provide opaque results, supporting informed decision-making.

[0134] In some embodiments, the user interface module 110 is configured to present the matching score and explainable insights indicating a contribution of each of the plurality of factors to the matching score. These explainable insights enable users to understand how the compatibility evaluation was derived and provide transparency in the decision-making process.

[0135] The explainable insights may include a breakdown of the plurality of factors used in the scoring, trend visualizations based on outputs generated by trend analysis and referencebased verification results generated by the reference chain verification module. The system 100 aggregates these outputs to provide a clear and interpretable representation of how individual factors and metrics influence the final compatibility score.

[0136] In some embodiments, the system 100 preferably enables the first entity 10 and the second entity 20 to modify data associated with the plurality of factors through an interactive interface. Upon such modification, the multi-factor scoring module 108 recalculates the matching score in real time or near real time, and the explainable insights are correspondingly updated. The updated insights may present the relative contribution and weighting of each factor in a clear and interpretable format, such as visual indicators, ranked lists, or comparative summaries. This dynamic updating mechanism allows users to understand how changes to input data influence the matching outcome, thereby improving transparency and decisionmaking.

[0137] The user interface module 110 may include score breakdown dashboards. These dashboards display the individual contributions of factors such as domain knowledge, functional skills, experience level, personality traits, temporal weighting and reference verification metrics. Each factor may be represented numerically and visually, enabling users to quickly identify the elements that most strongly influence the matching score.

[0138] The user interface module 110 may further present comparison charts that allow users to compare the attributes of the first entity 10 and the second entity 20 across multiple dimensions. Charts may illustrate alignment between domains, hierarchies, functions, values, motivations or other factors. These comparisons provide users with a comprehensive overview of compatibility across different aspects of the profiles of the entities 10, 20.

[0139] Another component of the user interface module 110 may be factor influence visualization. The user interface module 110 may graphically represent the relative influence of each factor on the final matching score. Factors with higher weighting may be highlighted, while factors with lower weighting may be displayed with reduced emphasis. This allows users to quickly identify potential areas for improvement in the compatibility assessment.

[0140] Factor influence visualizations may also illustrate how certain factors evolve over time. For example, charts may show progression of experience, development of skills or changes in reference verification metrics. These temporal insights provide additional context for understanding the matching score and may support strategic decision-making by users.

[0141] The explainable insights provided by the user interface module 110 enhance transparency, trust and interpretability in the matching process. By clearly presenting how each factor contributes to the overall score, the user interface module 110 allows users to make informed decisions based on both quantitative metrics and contextual explanations.Analytics Module

[0142] In some embodiments, the system 100 may further comprise an analytics module (not shown in FIG. 1) configured to aggregate and analyse the plurality of factors, the trend analysis outputs and the reference-based verification results and to generate analytical reportsand performance metrics derived from the matching score and the plurality of factors. The analytics module operates in conjunction with the multi-factor scoring module 108 to process both current and historical data associated with the first entity 10 and the second entity 20.

[0143] The analytics module may generate various performance metrics that evaluate compatibility strength, factor alignment and patterns associated with successful matching outcomes. These metrics may be used to assess the effectiveness of the matching process and to identify factors that most strongly contribute to favourable compatibility scores.

[0144] In embodiments such as HR matching, the analytics module can perform talent pipeline analysis by analysing aggregated data across multiple candidates and roles. This analysis may identify emerging skill distributions, domain- specific talent availability or gaps in expertise within a candidate pool.

[0145] The analytics module may also generate hiring outcome predictions or other forward-looking assessments by analysing relationships between factor patterns, reference verification results and past matching outcomes. In addition, the system 100 may produce trend reports that identify changes over time in factor importance, domain demand or compatibility patterns across entities 10, 20.

[0146] By aggregating and analysing matching data, the analytics module enables the system to transform individual compatibility evaluations into broader insights that support strategic decision-making and long-term planning.Data Secrecy and Privacy Implementation Mechanisms

[0147] In some embodiments, mechanisms may be provided to protect the confidentiality, integrity and controlled use of data associated with the first entity 10 and the second entity 20. Although data secrecy mechanisms are not required in all implementations, certain embodiments preferably incorporate privacy-preserving techniques that operate alongside the data acquisition module 106, the multi-factor scoring module 108 and the user interface module 110, as well as with the reference chain verification module and the analytics module ifapplicable. These mechanisms may be implemented as optional system features designed to safeguard sensitive information processed during compatibility evaluation.

[0148] In some embodiments, the system 100 may employ encryption techniques to protect data during transmission and storage. Data collected by the data acquisition module 106 may be encrypted prior to transmission across the network environment and encrypted storage may be used within the database infrastructure. Encryption may be applied to profile information, factor values, reference verification data and historical matching results. Communication between system components, including the web server, database access layer and scoring modules, may occur through secure communication protocols that prevent unauthorized interception or modification of data.

[0149] The system 100 may further implement differential privacy mechanisms to further protect sensitive data. Differential privacy techniques may be applied when generating or sharing outputs derived from matching computations. For example, controlled statistical noise may be introduced into aggregated outputs, compatibility metrics or analytical reports before such information is presented to external parties. By introducing carefully calibrated noise, the system 100 reduces the possibility that individual data attributes associated with the first entity 10 or the second entity 20 can be inferred from the reported results while preserving the overall utility of the matching insights.

[0150] The system 100 may further utilize federated learning techniques to support privacy-preserving model improvement. Under this approach, analytical models or weighting mechanisms associated with the multi-factor scoring module 108 may be trained across multiple distributed datasets without requiring raw data to be transferred to a centralized server. Instead, local model updates may be generated within separate data environments and subsequently aggregated to improve the overall model performance.

[0151] The system 100 may further incorporate network security mechanisms, such as firewall protection, intrusion detection systems and secure gateway services. These protections may be implemented within the network infrastructure that supports the web server, application servers and database servers. Firewall systems may regulate inbound and outbound communication to ensure that only authorized requests are permitted to interact with the systemcomponents. Additional monitoring mechanisms may detect suspicious access attempts or abnormal data access patterns.

[0152] By integrating encryption, differential privacy, federated learning, network security infrastructure and indexing control mechanisms, the system 100 may provide enhanced protection for sensitive information associated with compatibility evaluation.Implementation Environment

[0153] In some embodiments, the system 100 may be implemented within a networked computing environment comprising one or more computing devices, servers, communication networks, and data storage systems configured to support execution of the modules described above. One or more processors of the computing devices may execute computer-readable instructions stored in a memory or non-transitory storage medium to perform the abovedescribed processes and functions.

[0154] The system 100 may operate within a distributed computing architecture in which multiple computing nodes cooperate to process requests, perform matching computations, and manage stored data. The system 100 may be deployed on cloud-based or on-premises servers providing scalable computing resources. Such servers may dynamically allocate processing power, memory, and storage capacity in response to system demand, thereby enabling efficient execution of matrix operations, multi-factor scoring, and trend analysis processes.

[0155] In some embodiments, the functionality of the system 100 may be implemented using software, hardware, or a combination thereof. The system 100 may be realised as a computer program product comprising instructions which, when executed by one or more processors, cause the processors to perform the methods described herein. The instructions may be stored in a non-transitory computer-readable storage medium, including but not limited to random access memory (RAM), read-only memory (ROM), solid-state storage devices, magnetic storage devices, or other machine-readable media.

[0156] Communication between modules of the system 100 may occur through APIs, which provide standardized communication protocols for secure and structured data exchange.For example, the data acquisition module 106 may transmit collected data to a database via an API, while the multi-factor scoring module 108 may retrieve and process the data through corresponding API calls. In embodiments involving automated reference chain verification, APIs may further enable communication with external verification systems or trusted data providers.

[0157] The system 100 may further include a database access layer that provides an abstraction between system modules and underlying data storage technologies. The database access layer enables system components to store, retrieve, update, and query data without requiring direct knowledge of the database schema. The underlying database may include relational databases, distributed databases, or other structured or semi- structured storage systems. Stored data may include profile data, factor matrices, weighting parameters, reference verification data, historical matching scores, and analytics outputs generated during execution of the method (200).

[0158] User interaction with the system 100 may be facilitated through one or more interface modules, including web-based interfaces and mobile interfaces. A web server may manage communication between client devices and system components, process user requests, manage user sessions, and deliver interface content. Through such interfaces, users may input data via questionnaires, modify factor-related information, review matching scores, and access analytical outputs and trend visualizations.

[0159] In some embodiments, the system 100 may support execution across multiple smart device types, including desktop computers, laptops, smartphones, and tablets. Client devices may communicate with the system via wired or wireless networks, including the Internet, using secure communication protocols.

[0160] The implementation environment may further include security mechanisms for protecting sensitive data. These mechanisms may include encrypted communication channels, authentication and authorization protocols, firewall systems, intrusion detection systems, and data obfuscation techniques integrated with the data secrecy preservation module. Such mechanisms ensure that data remains protected during storage, processing, and transmission.

[0161] Accordingly, the implementation environment provides a computing framework in which one or more processors execute stored instructions to implement the matching methods described herein. The system 100 may therefore be realised as a computer-implemented method, an apparatus comprising processing circuitry and memory, a non-transitory computer-readable storage medium storing executable instructions, or a computer program product.

[0162] As described above, the present application implements a Meta-Matrix dictionary, which systematically categorizes candidate and employer metadata into hierarchical, domainspecific frameworks, including, but not limited to, domain, hierarchies, and functions. In contrast to static keyword-based systems, this taxonomy advantageously encodes semantic relationships among attributes, thereby enabling the system 100 to recognize conceptual equivalencies (for example, equivalence between “Project Manager” and “Agile Lead”) through contextual and relational analysis. Each attribute is assigned a meta- signature that advantageously quantifies its semantic depth, contextual variability, and relational dependencies within the taxonomy.

[0163] The system 100 and method (200) utilize a distinct and comprehensive set of elements encompassing all factors relevant to a given situation, person, or job, thereby ensuring thorough consideration of every aspect. The comprehensive set of factors advantageously possesses the following properties: inclusivity, whereby all significant factors are incorporated with no material omissions; broad scope, whereby a wide array of elements across multiple dimensions is addressed; and comprehensiveness, whereby the entirety of pertinent information is considered to enable holistic analysis.

[0164] The present application further identifies and incorporates relevant market trends, industry dynamics, and external factors to ensure that entities are advantageously matched with opportunities aligned with future demands. Dimensions across physical, temporal, and virtual realms are advantageously incorporated into the scoring mechanisms based on such trend analysis. The system 100 and method (200) considers multiple types of dimensions, including all-encompassing, multi-domain dimensions relating to systems and structures. This includes multi-systems that advantageously integrate expertise from diverse fields such as medicine, banking, and education.

[0165] Furthermore, the system 100 and method (200) utilize set theory by treating entity factors as sets, thereby advantageously enabling sophisticated matching based on set intersections, unions, and differences. To enhance robustness, the system 100 incorporates functionality for considering unknowns, i.e., aspects of an entity that are not explicitly stated but can be inferred from available data or external sources.

[0166] The system 100 and method (200) operate as a multi-domain, multi-function, and multi-role system and method, incorporating question sets tailored for various industries, including medical, IT, education, and others. This ecosystem of multi-domain questions and dimensions for each trait advantageously enables more accurate and relevant assessments across different sectors. For example, in the medical domain, questions may focus on patient care, historical performance, expertise in specific medical procedures, and adherence to healthcare regulations. In IT, the system 100 and method (200) may emphasize technical capabilities, project management historical performance, and familiarity with emerging technologies. In education, questions may focus on teaching methodologies, curriculum development, and student engagement strategies.

[0167] This multi-domain-specific approach advantageously ensures that the system 100 and method (200) can accurately evaluate candidates and individuals across a wide range of industries, thereby providing a more nuanced and contextually relevant assessment. The dimensions for each trait are advantageously tailored to reflect the unique requirements and challenges of each domain, enabling a more comprehensive evaluation of a target’s suitability for a particular role within a specific field.

[0168] In an HR application for example, the system 100 and method (200) advantageously enable employers to apply multi-layered filters to selectively eliminate or identify candidates based on nuanced, domain- specific criteria. For instance, an employer may filter candidates by specifying attributes such as “Hierarchies”, “Strategist and Functions” and “Business Analysis”. This action triggers a meta-query that identifies candidates possessing correlated attributes, such as visionary character traits or demonstrated strategic planning experience, while excluding candidates with semantically unrelated profiles. The fine-grained filtering capability advantageously prevents excessive narrowing of candidate pools and preserves nuanced, contextually relevant matches.

[0169] In the same HR application, the system 100 and method (200) further incorporate demographic-occupational elimination mechanisms that assess alignment between demographic factors and legitimate occupational requirements. Unlike conventional demographic filtering approaches that may raise concerns regarding discriminatory practices, the present system 100 and method (200) advantageously distinguish between impermissible preferences and genuine job-related criteria, thereby maintaining compliance with equal employment opportunity regulations.

[0170] Preferably, the system 100 and method (200) can evaluate whether a given demographic-related filter corresponds to a bona fide occupational requirement. For example, the system 100 and method (200) can advantageously distinguish between a preference for a particular age range, which may be discriminatory, and a legitimate requirement related to physical capabilities necessary for specific roles, which may be legally permissible. This distinction is achieved through contextual analysis of role requirements, domain- specific standards, and regulatory considerations, thereby ensuring that filtering decisions remain both relevant and compliant.

[0171] While there has been described in the foregoing description exemplary embodiments of the present invention, it will be understood by those skilled in the technology concerned that many variations in details of design, construction and / or operation may be made without departing from the present invention. It will be appreciated that many further alterations, modifications and permutations of various aspects of the described embodiments are possible that fall within the spirit and scope of the appended claims.

[0172] Throughout this specification and the claims which follow, unless the context requires otherwise, the word “comprise” and variations such as “comprises” or “comprising”, will be understood to imply the inclusion of a stated integer or group of integers or steps but not the exclusion of any other integer or group of integers. The reference in this specification to any prior publication (or information derived from it) or to any matter which is known, is not and should not be taken as an acknowledgment or admission or any form of suggestion that that prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of endeavour to which this specification relates.

Claims

Claims1. A computer-implemented method for evaluating compatibility between a first entity and a second entity, the method comprising:obtaining data associated with the first entity and data associated with the second entity; deriving a plurality of factors from the obtained data, wherein the plurality of factors comprise at least one factor representing an attribute of the first entity and at least one factor representing an attribute of the second entity;performing matrix operations that combine values of the plurality of factors and weights associated with the plurality of factors across dimensions and hierarchical levels to compute a matching score between the first entity and the second entity; andpresenting the matching score and explainable insights indicating a contribution of each of the plurality of factors to the matching score.

2. The method of claim 1, further comprising dynamically adjusting the weights in response to changes in values of the plurality of factors associated with at least one of the first entity and the second entity.

3. The method of claim 1 or 2, further comprising performing spectrum analysis on the plurality of factors to determine the weights for determining the matching score.

4. The method of any one of claims 1 to 3, further comprising processing the plurality of factors in a defined sequence based on an order of entry of the data, wherein the sequence is adjustable to influence the resulting matching score.

5. The method of any one of claims 1 to 4, further comprising including relationships among the plurality of factors based on hierarchical groupings of the factors in the matrix operations when determining the matching score.

6. The method of any one of claims 1 to 5, further comprising applying temporal weighting to the plurality of factors such that an influence of each factor on the matching score varies depending on at least one of a historical state, a current state and a predicted state of at least one of the first entity and the second entity.

7. The method of any one of claims 1 to 6, further comprising evaluating the plurality of factors in an acyclic manner.

8. The method of any one of claims 1 to 7, further comprising selecting different subsets of the plurality of factors for inclusion in computing the matching score.

9. The method of any one of claims 1 to 8, further comprising allowing the first entity and the second entity to modify data associated with the plurality of factors used in computing the matching score.

10. The method of any one of claims 1 to 9, further comprising:obtaining reference data from a plurality of reference sources associated with at least one of the first entity and the second entity;evaluating the reference data to generate a reference-based verification output; and incorporating the reference-based verification output into the matching score.

11. The method of claim 10, wherein the explainable insights comprise: a breakdown of the plurality of factors used in the scoring, trend visualizations based on trend analysis outputs and reference-based verification results.

12. The method of claim 11, further comprising:aggregating and analysing the plurality of factors, the trend analysis outputs and the reference-based verification results; andgenerating at least one of analytical reports and performance metrics derived from the matching score and the plurality of factors.

13. A computer-implemented matching system for evaluating compatibility between a first entity and a second entity, the system comprising:a data acquisition module configured to obtain data associated with the first entity and data associated with the second entity;a multi-factor scoring module configured to derive a plurality of factors from the obtained data, wherein the plurality of factors comprise at least one factor representing an attribute of the first entity and at least one factor representing an attribute of the second entityand to perform matrix operations that combine values of the plurality of factors and weights associated with the plurality of factors across multiple dimensions and hierarchical levels to compute a matching score between the first entity and the second entity; anda user interface module configured to present the matching score and explainable insights indicating a contribution of each of the plurality of factors to the matching score.

14. The system of claim 13, wherein the multi-factor scoring module is further configured to:dynamically adjust the weights in response to changes in values of the plurality of factors; andperform spectrum analysis on the plurality of factors to determine the weights for determining the matching score.

15. The system of claim 13 or 14, wherein the multi-factor scoring module is further configured to:process the plurality of factors in a defined sequence based on an order of entry of the data, wherein the sequence is adjustable to influence the resulting matching score;include relationships among the plurality of factors based on hierarchical groupings of the factors in the matrix operations when determining the matching score;apply temporal weighting to the plurality of factors such that an influence of each factor on the matching score varies depending on at least one of a historical state, a current state and a predicted state of at least one of the first entity and the second entity; andevaluate the plurality of factors in an acyclic manner.

16. The system of any one of claims 13 to 15, wherein the multi-factor scoring module is further configured to:select different subsets of the plurality of factors for inclusion in computing the matching score; andallow the first entity and the second entity to modify data associated with the plurality of factors used in computing the matching score; andwherein the system further comprises a reference chain verification module configured to obtain reference data from a plurality of reference sources, generate a reference-based verification output and provide the reference-based verification output to the multi-factor scoring module for incorporation into the matching score.

17. The system of any one of claims 13 to 16, wherein the explainable insights comprise a breakdown of the plurality of factors, trend visualizations based on trend analysis outputs and reference-based verification results; andwherein the system further comprises an analytics module configured to aggregate and analyse the plurality of factors, the trend analysis outputs and the reference-based verification results and generate at least one of analytical reports and performance metrics derived from the matching score and the plurality of factors.

18. A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform the method of any one of claims 1 to 12.

19. An apparatus comprising one or more processors and a memory storing instructions that, when executed by the one or more processors, cause the apparatus to perform the method of any one of claims 1 to 12.

20. A computer program product, wherein the computer program product comprises instructions for performing the method according to any one of claims 1 to 12.

Images