Expert extraction system and method for science and technology reward review

Abstract

The invention relates to an expert extraction system for science and technology reward review. The expert extraction system comprises an evaluation-participating enterprise information database, an evaluation-participating expert information database, an enterprise social relation database, an expert social relation database, a random extraction module, an avoidance module and a result display module. According to the method, by optimizing the short-term frequency imbalance problem of the pure random model, the expert reference opportunity is more balanced, and the benefit association experts are intercepted in real time in combination with the dynamic relation graph, so that the extraction fairness is ensured; the mode of self-service extraction and multi-factor authentication is adopted, the tedious links of repeated confirmation and paper signature of traditional manual extraction are broken through, the whole process is in online closed-loop operation, and the operation time is shortened by 90% or above.

Description

Technical Field

[0001] This invention relates to the field of extraction system technology, and in particular to an expert extraction system and method for reviewing science and technology awards. Background Technology

[0002] With the increasing frequency of IT project applications and reviews, management departments at all levels urgently need to improve the efficiency and impartiality of expert reviews through intelligent means. As a core component of project quality control, the accuracy of expert reviews directly impacts resource allocation and the scientific nature of decision-making.

[0003] Traditional manual selection methods and existing electronic random selection methods are insufficient to meet the needs of managing large-scale expert databases and are easily influenced by subjective factors, failing to meet the mandatory requirements for fair and transparent processes.

[0004] Existing expert extraction technologies are limited by algorithmic simplification, fragmented data, and static regulatory oversight, making it difficult to meet the demands of modern review processes that require high transparency, strong constraints, and full automation. Therefore, there is an urgent need to develop a novel extraction system that uses rigid technological constraints to address fairness and security vulnerabilities, thereby driving the review process towards standardization and intelligence. Summary of the Invention

[0005] The main objective of this invention is to provide an expert selection system and method for reviewing science and technology awards, thereby addressing the deficiencies of existing technologies.

[0006] To achieve the above objectives, the specific plan is as follows:

[0007] An expert selection system for reviewing science and technology awards includes: a database of information on participating enterprises, a database of information on participating experts, a database of enterprise social relations, a database of expert social relations, a random selection module, an avoidance module, and a results display module;

[0008] The database of participating companies' information is used to record and retrieve the names, taxpayer identification numbers, technical fields, and brief descriptions of the technical content of participating companies.

[0009] The database of participating experts records the names, ID numbers, total number of participating experts, and their technical fields.

[0010] A corporate social relations database is used to record information on corporate shareholders, corporate management personnel, related information on corporate shareholders, related information on corporate management personnel, and related information on external experts hired by the company.

[0011] The expert social relations database is used to record the current employment status and past employment status of the participating experts.

[0012] The random selection module is used to select evaluation experts for different participating companies;

[0013] The avoidance module is used to determine the pool of participating experts based on the enterprise social relations database and the expert social relations database, in conjunction with the random sampling module.

[0014] The results display module is used to ultimately display the pool of experts for each round of projects and participating companies.

[0015] Furthermore, the expert extraction method for querying science and technology awards includes:

[0016] The S1 random selection module, based on the participating companies' technical fields, technical content, and the experts' technical fields, randomly selects N candidate experts from the expert information database. The set of candidate experts is defined as C = e1, e2, ..., e N The set of candidate companies formed by the participating companies in this round of evaluation is defined as EntP = ent1, ent2, ..., ent K The number of times each expert participates in the evaluation is defined as C. i The social relationship matrix is ​​defined as ESR = [esr] ij ] N×N ,esr ij ∈[0,1] represents expert e i With e j society

[0017] Relationship strength;

[0018] ETR = [etr] ik ] N×K ,etr ik ∈0,1 represents expert e i With enterprise ent k Relationship (1 for existence)

[0019] (related);

[0020] The S2 random sampling module calculates the equilibrium weight based on the number of times experts participate in the evaluation, using the following formula:

[0021]

[0022] The S3 random sampling module calculates constraint values ​​based on the expert social relationship database and the enterprise social relationship database, as follows:

[0023] (3) Expert relationship avoidance density: Define the social relationship density constraint within the expert set, as follows:

[0024] (4) Firm association avoidance index, which defines the constraint of expert association with firm, and the formula is as follows:

[0025]

[0026] The S4 random sampling module calculates the dynamic balance value between random weights and social relationship constraints, using the following formula:

[0027]

[0028] The S5 random sampling module filters experts within the candidate expert set C based on the ETR (Enterprise Association Avoidance Index). If e i There is an association with the project-related companies (etr) ik If the value is 1), then it is directly excluded;

[0030] The S6 random sampling module processes the expert set C filtered in the previous step according to the dynamic equilibrium value P. select (e i |S) Select suitable experts and determine whether the required number has been reached. If the number is not reached, repeat step S6 until the number is reached and output the final list of experts participating in this round.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. This invention optimizes the short-term frequency imbalance problem of the pure random model, making the opportunities for expert participation more balanced, and combines dynamic relationship graphs to intercept experts with conflicting interests in real time, ensuring the fairness of the selection.

[0033] 2. This invention adopts the "self-service extraction + multi-factor authentication" model, which breaks the cumbersome process of repeated confirmation and paper signature required by traditional manual extraction. The entire process is online and closed-loop, reducing the operation time by more than 90%. Attached Figure Description

[0034] Figure 1 This is a flowchart of the extraction process of the present invention; Detailed Implementation

[0035] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments.

[0036] Combination Figure 1 The aforementioned expert selection system for reviewing science and technology awards includes: a database of information on participating enterprises, a database of information on participating experts, a database of enterprise social relations, a database of expert social relations, a random selection module, an avoidance module, and a result display module.

[0037] The database of participating companies' information is used to record and retrieve the names, taxpayer identification numbers, technical fields, and brief descriptions of the technical content of participating companies.

[0038] The database of participating experts records the names, ID numbers, total number of participating experts, and their technical fields.

[0039] A corporate social relations database is used to record information on corporate shareholders, corporate management personnel, related information on corporate shareholders, related information on corporate management personnel, and related information on external experts hired by the company.

[0040] The expert social relations database is used to record the current employment status and past employment status of the participating experts.

[0041] The random selection module is used to select evaluation experts for different participating companies;

[0042] The avoidance module is used to determine the pool of participating experts based on the enterprise social relations database and the expert social relations database, in conjunction with the random sampling module.

[0043] The results display module is used to ultimately display the pool of experts for each round of projects and participating companies.

[0044] Furthermore, the expert extraction method for querying science and technology awards includes:

[0045] The S1 random selection module, based on the participating companies' technical fields, technical content, and the experts' technical fields, randomly selects N candidate experts from the expert information database. The set of candidate experts is defined as C = e1, e2, ..., e N The set of candidate companies formed by the participating companies in this round of evaluation is defined as EntP = ent1, ent2, ..., ent K The number of times each expert participates in the evaluation is defined as C. i The social relationship matrix is ​​defined as ESR = [esr] ij ] N×N ,esr ij ∈[0,1] represents expert e i With e j society

[0046] Relationship strength;

[0047] ETR = [etr] ik ] N×K ,etr ik ∈0,1 represents expert e i With enterprise ent k Relationship (1 for existence)

[0048] (related);

[0049] The S2 random sampling module calculates the equilibrium weight based on the number of times experts participate in the evaluation, using the following formula:

[0050]

[0051] The S3 random sampling module calculates constraint values ​​based on the expert social relationship database and the enterprise social relationship database, as follows:

[0052] (5) Expert relationship avoidance density: Define the social relationship density constraint within the expert set, as follows:

[0053] (6) Firm Association Avoidance Index: Defines the association constraints between experts and firms, as shown in the following formula:

[0054]

[0055] The S4 random sampling module calculates the dynamic balance value between random weights and social relationship constraints, using the following formula:

[0056]

[0057] The S5 random sampling module filters experts within the candidate expert set C based on the ETR (Enterprise Association Avoidance Index). If e i There is an association with the project-related companies (etr) ik If the value is 1), then it is directly excluded;

[0059] The S6 random sampling module processes the expert set C filtered in the previous step according to the dynamic equilibrium value P. select (e i |S) Select suitable experts and determine whether the required number has been reached. If the number is not reached, repeat step S6 until the number is reached and output the final list of experts participating in this round.

[0060] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An expert selection system for reviewing science and technology awards, characterized in that, include: The system includes a database of participating companies' information, a database of participating experts' information, a database of companies' social relations, a database of experts' social relations, a random selection module, an avoidance module, and a results display module. The database of participating companies' information is used to record and retrieve the names, taxpayer identification numbers, technical fields, and brief descriptions of the technical content of participating companies. The database of participating experts records the names, ID numbers, total number of participating experts, and their technical fields. A corporate social relations database is used to record information on corporate shareholders, corporate management personnel, related information on corporate shareholders, related information on corporate management personnel, and related information on external experts hired by the company. The expert social relations database is used to record the current employment status and past employment status of the participating experts. The random selection module is used to select evaluation experts for different participating companies; The avoidance module is used to determine the pool of participating experts based on the enterprise social relations database and the expert social relations database, in conjunction with the random sampling module. The results display module is used to ultimately display the pool of experts for each round of projects and participating companies.

2. The expert selection method for science and technology award inquiry according to claim 1, characterized in that, include: The S1 random selection module, based on the participating companies' technical fields, technical content, and the experts' technical fields, randomly selects N candidate experts from the expert information database. The set of candidate experts is defined as C = e1, e2, ..., e N The set of candidate companies formed by the participating companies in this round of evaluation is defined as EntP = ent1, ent2, ..., ent K The number of times each expert participates in the evaluation is defined as C. i The social relationship matrix is ​​defined as ESR = [esr] ij ] N×N ,esr ij ∈[0,1] represents expert e i With e j The strength of social relationships; ETR = [etr] ik ] N×K ,etr ik ∈0,1 represents expert e i With enterprise ent k The correlation (1 indicates that a correlation exists); The S2 random sampling module calculates the equilibrium weight based on the number of times experts participate in the evaluation, using the following formula: The S3 random sampling module calculates constraint values ​​based on the expert social relationship database and the enterprise social relationship database, as follows: (1) Expert relationship avoidance density: Define the social relationship density constraint within the expert set, as follows: (2) Firm Association Avoidance Index: The constraint on the association between experts and firms is defined by the following formula: The S4 random sampling module calculates the dynamic balance value between random weights and social relationship constraints, using the following formula: The S5 random sampling module filters experts within the candidate expert set C based on the ETR (Enterprise Association Avoidance Index). If e i There is an association with the project-related companies (etr) ik If the value is 1), then it is directly excluded; The S6 random sampling module processes the expert set C filtered in the previous step according to the dynamic equilibrium value P. select (e i |S) Select suitable experts and determine whether the required number has been reached. If the number is not reached, repeat step S6 until the number is reached and output the final list of experts participating in this round.

Images