A data intelligent analysis system and method of a science and technology management platform

Abstract

The application discloses a data intelligent analysis system and method of a science and technology management platform, relates to the technical field of project matching analysis, and comprises the following steps: collecting data in a science and technology enterprise research and development management platform, performing cleaning, deduplication and standardization processing, and establishing associated relationship data between research and development personnel and projects; constructing a personnel capability-project progress multi-dimensional evaluation index system; according to the multi-dimensional evaluation index system, analyzing historical project data and achievement data of the research and development personnel, calculating a research and development personnel capability characteristic vector, and constructing a research and development capability portrait model; based on the research and development capability portrait model and technical demand characteristics of a current project, calculating a matching relationship between the research and development personnel and the project, obtaining a team-project matching degree score, and generating a research and development team configuration suggestion. The application improves overall research and development resource configuration efficiency through comprehensive evaluation of candidate teams and recommendation of optimal teams.

Description

Technical Field

[0001] This invention relates to the field of project matching analysis technology, specifically a data intelligent analysis system and method for a technology management platform. Background Technology

[0002] As R&D activities in technology companies become increasingly complex, R&D projects typically involve collaboration across multiple technology fields, multi-stage advancement, and participation from multiple roles. How to achieve the rational allocation and efficient management of R&D resources has become a key issue for companies to enhance their innovation capabilities.

[0003] While existing enterprise R&D management platforms can perform basic management of project progress, personnel information, and results data, they still have significant shortcomings in data utilization, mainly in the following aspects: On the one hand, existing systems mostly evaluate the R&D process using single-dimensional indicators, lacking a detailed portrayal of the R&D personnel's capabilities in different technical fields, and thus failing to truly reflect the comprehensive technical level of the R&D personnel; On the other hand, existing technologies generally lack a mechanism for integrating and analyzing the quality of project phase deliverables, personnel contributions, and project progress, resulting in one-sided project evaluation results that cannot provide effective support for R&D decisions. In addition, during the process of building R&D teams, the allocation of personnel relies heavily on human experience and lacks a data-driven matching model. This makes it difficult to achieve a precise match between the capabilities of R&D personnel and project requirements, which can easily lead to resource mismatch or capability redundancy. Furthermore, in the case of multiple projects running in parallel, existing technologies lack a unified optimization mechanism to comprehensively evaluate multiple candidate teams, making it difficult to achieve optimal decision-making in overall resource allocation.

[0004] Therefore, there is an urgent need for an intelligent analysis method and system that can integrate multi-source data, construct a multi-dimensional evaluation index system, and realize the modeling of R&D personnel capabilities and the optimization of project matching, so as to improve the scientific and intelligent level of R&D management. Summary of the Invention

[0005] The purpose of this invention is to provide a data intelligent analysis system and method for a technology management platform to solve the problems raised in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a data intelligent analysis method for a technology management platform, specifically including the following steps, S1. Collect project data, R&D personnel data, and technological achievement data from the R&D management platform of technology companies, clean, deduplicate, and standardize them, and establish the relationship data between R&D personnel and projects based on personnel participation relationships; S2. Based on the project data, R&D personnel data, and technological achievement data, construct a multi-dimensional evaluation index system of personnel capability and project progress. The personnel capability index is used to characterize the technical capability level of R&D personnel in different technical fields, and the project progress index is used to characterize the degree of task completion of the R&D project at different stages. S3. Based on the multidimensional evaluation index system, analyze the historical project participation data and achievement data of R&D personnel, calculate the capability feature vector of R&D personnel, and construct an R&D capability profile model. S4. Based on the R&D capability profile model and the technical requirements of the current project, calculate the matching relationship between R&D personnel and the project, obtain the team-project matching score, and generate R&D team configuration suggestions.

[0007] In S1, project data, R&D personnel data, and technological achievement data are collected from the R&D management platform of technology companies. This data is then cleaned, deduplicated, and standardized. Furthermore, data linking R&D personnel to projects is established based on their involvement. Specifically: Identify the data source systems within the enterprise that are involved in R&D management, obtain data access tokens through the enterprise's unified identity authentication mechanism, match user role permissions based on the access tokens, generate corresponding data access control policies, and authorize access to relevant data in the project management system, human resources system, and technology achievement management system while ensuring data security. After authorization, data related to R&D activities are collected from various data source systems. Project data includes at least the project number, project name, technical field, project stage status, and task breakdown information; R&D personnel data includes at least a unique personnel identifier, technical field tag, team affiliation, and historical project participation records; and technical achievement data includes at least technical document information and stage achievement reports. During the data collection process, timestamps and data source identifiers are added to all types of data to ensure the traceability of data sources.

[0008] Subsequently, the collected data underwent data quality processing, including detecting and completing missing fields or marking them as outliers, identifying duplicate data through key field comparison and retaining the latest data version, and detecting outliers in fields such as project cycle and number of deliverables to improve the accuracy and reliability of the data.

[0009] After data cleaning is completed, data from different systems are standardized and uniformly processed, including uniform field naming rules, uniform data units, and standardized mapping of technical field labels, in order to eliminate differences in data structure between different systems. Based on the records of R&D personnel's participation in projects, establish the correlation data between R&D personnel and projects, and record the personnel's participation role, participation time, and task contribution information in the correlation data. Construct a triple of participation role, participation time, and task contribution information as the correlation data; provide basic data support for subsequent R&D personnel capability analysis and intelligent analysis of project data.

[0010] In S2, based on the project data, R&D personnel data, and technological achievement data, a multi-dimensional evaluation index system of personnel capability and project progress is constructed. The personnel capability index characterizes the technical capability level of R&D personnel in different technical fields, and the project progress index characterizes the degree of task completion in different stages of the R&D project. Specifically, this includes the following steps: S2-1: Based on the project's phase status, construct a project deliverable quality assessment model, specifically characterized as follows: Q p =∑ k=1 ns w k *v k ; where Q p This represents the quality score of the R&D personnel for each stage of the project; s represents the stage of the project; ns represents the total number of deliverable types included in stage s; k represents the type of deliverables in the stage; v k Scoring of interim results; k Weights for different types of stage deliverables; S2-2: Based on collected project data, R&D personnel data, and technological achievement data, quantitatively evaluate the capabilities of R&D personnel in different technical fields to form personnel capability indicators; the specific method is as follows: statistically analyze the task contribution and stage achievement quality score of each R&D personnel in the project set they participate in, and calculate the personnel's capability index (TAI) in the technical field t. i,t : Where i represents R&D personnel; p represents projects; C i,p This indicates the researcher's contribution to project p; P i t Represents the set of items in the technical field t; S2-3: Based on project task breakdown information, milestone data, and R&D activity records, calculate project progress indicators, represented as: PPI. p =∑ s=1 S w s *(T completed,s / T total,s ); among which, PPI p T represents the project progress indicators for project p; completed,s Indicates the amount of task completed in stage s; T total,s w represents the total amount of work in phase s.s The stage weights are S; S represents the total number of stages in project p. S2-4: Normalize personnel capability indicators and project progress indicators: Unify indicators of different dimensions to a scale range of 0-1, forming a multi-dimensional evaluation vector V. p,s Characterized as: V p,s =[TAI i,t * PPI p ]; Among them, TAI i,t * This represents the average value of an index indicating an individual's competence in the technical field.

[0011] In S3, based on the aforementioned multidimensional evaluation index system, the historical project participation data and achievement data of R&D personnel are analyzed to calculate the R&D personnel capability feature vector and construct an R&D capability profile model. Specifically, this includes the following steps: S3-1: Based on S2, a multi-dimensional evaluation index system for personnel capability and project progress is constructed. This system analyzes the historical project data of R&D personnel, extracts the task contribution, role, quality of project stage results, and project completion status of each R&D personnel in each project, and classifies the projects according to the technical field, thereby obtaining a set of R&D personnel participation records in different technical fields. S3-2: Arrange the R&D personnel's capability indices across all technical fields according to the preset technical field set T={t1,t2,…,tm}, and construct the R&D personnel capability feature vector Fi: Fi=[TAI i,t1 TAI i,t2 ,…,TAI i,tm ]; where t1, t2, ..., tm represent the preset 1st, 2nd, and mth technical fields; TAI i,t1 TAI i,t2 ,…,TAI i,tm This represents the capability index of researcher i in the 1st, 2nd, and mth technical fields; S3-3: Based on the aforementioned R&D personnel capability feature vector, and integrating the historical output quantity and project execution performance of R&D personnel, according to the R&D personnel's project output quality index Q... p Project progress indicator PPI p Calculate the comprehensive ability score (CA) of R&D personnel. i : CA i =α*(Σ p∈Pi Q p ) / |P i |+β*(Σ p∈Pi PPI p) / |P i |; Among them, P i Let represent the set of projects in which R&D personnel i participates; α and β are weighting coefficients, satisfying α+β=1; S3-4: Combine the R&D personnel's capability feature vector Fi with the R&D personnel's comprehensive capability score CA i Combine the vectors to form a comprehensive capability representation vector Ai for R&D personnel: Ai=[Fi,CAi]; calculate the similarity between different comprehensive capability representation vectors of R&D personnel, perform capability clustering analysis on R&D personnel, and construct an R&D capability profile model; among them, the capability similarity between R&D personnel i1 and R&D personnel i2 is calculated by cosine similarity.

[0012] In S4, based on the R&D capability profile model and the technical requirements of the current project, the matching relationship between R&D personnel and the project is calculated to obtain a team-project matching score, and R&D team configuration suggestions are generated. Specifically, this includes the following steps: S4-1: Extract the technical requirements features of the target project for the R&D team to be configured; based on the project's technical solution description, task decomposition information, and project technical field tags, identify the set of technical fields Tp involved in the project. Tp = {t1, t2, ..., tm1}; t1, t2, tm1 represent the 1st, 2nd, and m1th technical fields involved in project p, respectively; Tp ⊆ T; Demand weights are assigned based on the importance of different technical fields in the project tasks, constructing a project technical demand vector. The technical demand vector Rp of project p is represented as: Rp = [w p,t1 ,w p,t2 ,…,w p,tm1 ];w p,t1 ,w p,t2 ,…,w p,tm1 Let w represent the demand weights of technical fields t1, t2, and tm1 in project p, respectively, satisfying: p,t1 +w p,t2 +…+w p,tm1 =1; S4-2: Using the R&D capability profile constructed in step S3, retrieve the capability indices for the 1st, 2nd, and m1th technical fields involved in project p, and extract the capability feature vectors Fi,p for each R&D personnel. Fi,p=[TAI i,t1 TAI i,t2 ,…,TAI i,tm1 By matching and analyzing the capability feature vector of R&D personnel with the technical requirement vector of the project, the capability matching degree r between R&D personnel i and project p is calculated: This is represented as: r=(Fi,p∙Rp) / (∥Fi,p∥∙∥Rp∥); where ∥Fi,p∥ and ∥Rp∥ represent the norms of the corresponding vectors, respectively. S4-3: Based on the capability matching degree between R&D personnel i and project p and the team size required for the project, R&D personnel whose capability matching degree r is greater than a preset matching threshold are added to the candidates to generate the candidate R&D team G for project p.

[0013] Among these, as a preferred option, recommending the optimal R&D team for the N target projects currently requiring R&D team configuration includes the following steps: Step S1: Extract technical requirement features for each of the N target projects of the current R&D team to be configured; identify the technical fields involved in the project based on the project technical solution description, task decomposition information and project technical field tags; set requirement weights according to the importance of different technical fields in the project tasks; sort the N target projects in descending order based on the technical field with the highest requirement weight, and denote it as the target project sequence Y; Step S2: Based on the methods in S4-2 to S4-3, generate D candidate R&D teams for the target project; using the technical field of the target project sequence Y as the benchmark, sort the R&D personnel according to the capability index of the technical field from largest to smallest to obtain the R&D personnel capability feature sequence corresponding to the project sequence. The R&D personnel capability feature sequence of the D candidate R&D teams corresponding to the target project sequence Y is denoted as X=[X1,X2,…,Xd,…,XD], where Xk represents the R&D personnel capability feature sequence of the d-th candidate R&D team. Step S3: For any target project in the target project sequence Y involving h technical fields, each candidate R&D team shall include at least h R&D personnel. The capability index of R&D personnel in the same technical field shall be represented by the average value. The correlation ρ between the target project sequence and the capability characteristic sequence of R&D personnel of the candidate R&D team shall be calculated by the Pearson correlation coefficient method. The candidate R&D team with the largest |ρ| shall be recommended as the optimal R&D team.

[0014] A data intelligence analysis system for a technology management platform includes a data acquisition and processing module, an indicator construction module, a capability profile construction module, and a project matching analysis module; The data acquisition and processing module is used to acquire R&D-related data from multiple business systems, and to complete data preprocessing and correlation modeling, providing a unified data foundation for subsequent analysis; The indicator construction module is used to construct a multi-dimensional evaluation indicator system for personnel capabilities and project progress, so as to realize the quantitative expression of key indicators in the R&D process. The capability profile building module is used to model the capabilities of R&D personnel in multiple dimensions to form an R&D capability profile model; The project matching analysis module is used to perform matching analysis between R&D personnel and projects, and to generate candidate R&D teams.

[0015] Furthermore, it also includes a team optimization recommendation module; the team optimization recommendation module is used to recommend the optimal R&D team based on correlation analysis in multi-project scenarios.

[0016] The correlation analysis was performed using the Pearson correlation coefficient method.

[0017] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention unifies the collection, cleaning, and standardization of project data, R&D personnel data, and technological achievement data in the R&D management platform of technology enterprises, and establishes personnel-project correlation to achieve integrated management of multi-source data; it constructs a multi-dimensional evaluation index system based on personnel capabilities, project progress, and the quality of stage results, analyzes historical project participation data of R&D personnel to form capability feature vectors and constructs an R&D capability profile model, achieving a refined quantification and structured expression of R&D personnel capabilities; further, by calculating the matching relationship between R&D personnel capability feature vectors and project technical requirement vectors, it generates team-project matching scores and optimizes the selection of candidate R&D teams, achieving precise matching between R&D personnel and projects; simultaneously, in multi-project scenarios, it improves the overall efficiency of R&D resource allocation through comprehensive evaluation of candidate teams and recommendation of the optimal team. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a data intelligent analysis method for a technology management platform according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example: Figure 1 As shown, the present invention provides a technical solution, a data intelligent analysis method for a technology management platform, which specifically includes the following steps: S1. Collect project data, R&D personnel data, and technological achievement data from the R&D management platform of technology companies, clean, deduplicate, and standardize them, and establish the relationship data between R&D personnel and projects based on personnel participation relationships; In S1, project data, R&D personnel data, and technological achievement data are collected from the R&D management platform of technology companies. This data is then cleaned, deduplicated, and standardized. Furthermore, data linking R&D personnel to projects is established based on their involvement. Specifically: Identify the data source systems within the enterprise that are involved in R&D management, obtain data access tokens through the enterprise's unified identity authentication mechanism, match user role permissions based on the access tokens, generate corresponding data access control policies, and authorize access to relevant data in the project management system, human resources system, and technology achievement management system while ensuring data security. After authorization, data related to R&D activities are collected from various data source systems. Project data includes at least the project number, project name, technical field, project stage status, and task breakdown information; R&D personnel data includes at least a unique personnel identifier, technical field tag, team affiliation, and historical project participation records; and technical achievement data includes at least technical document information and stage achievement reports. During the data collection process, timestamps and data source identifiers are added to all types of data to ensure the traceability of data sources.

[0021] Subsequently, the collected data underwent data quality processing, including detecting and completing missing fields or marking them as outliers, identifying duplicate data through key field comparison and retaining the latest data version, and detecting outliers in fields such as project cycle and number of deliverables to improve the accuracy and reliability of the data.

[0022] After data cleaning is completed, data from different systems are standardized and uniformly processed, including uniform field naming rules, uniform data units, and standardized mapping of technical field labels, in order to eliminate differences in data structure between different systems. Based on the records of R&D personnel's participation in projects, establish the correlation data between R&D personnel and projects, and record the personnel's participation role, participation time, and task contribution information in the correlation data. Construct a triple of participation role, participation time, and task contribution information as the correlation data; provide basic data support for subsequent R&D personnel capability analysis and intelligent analysis of project data.

[0023] S2. Based on the project data, R&D personnel data, and technological achievement data, construct a multi-dimensional evaluation index system of personnel capability and project progress. The personnel capability index is used to characterize the technical capability level of R&D personnel in different technical fields, and the project progress index is used to characterize the degree of task completion of the R&D project at different stages. In S2, based on the project data, R&D personnel data, and technological achievement data, a multi-dimensional evaluation index system of personnel capability and project progress is constructed. The personnel capability index characterizes the technical capability level of R&D personnel in different technical fields, and the project progress index characterizes the degree of task completion in different stages of the R&D project. Specifically, this includes the following steps: S2-1: Based on the project's phase status, construct a project deliverable quality assessment model, specifically characterized as follows: Q p =∑ k=1 ns w k *v k ; where Q p This represents the quality score of the R&D personnel for each stage of the project; s represents the stage of the project; ns represents the total number of deliverable types included in stage s; k represents the type of deliverables in the stage; v k Scoring of interim results; k Weights for different types of stage deliverables; By weighting and summarizing the quality of deliverables at each stage, the technological value and innovation level of a project at different stages can be quantified.

[0024] S2-2: Based on collected project data, R&D personnel data, and technological achievement data, quantitatively evaluate the capabilities of R&D personnel in different technical fields to form personnel capability indicators; the specific method is as follows: statistically analyze the task contribution and stage achievement quality score of each R&D personnel in the project set they participate in, and calculate the personnel's capability index (TAI) in the technical field t. i,t : Where i represents R&D personnel; p represents projects; C i,p This indicates the researcher's contribution to project p; P i t This represents the set of projects in the technology field t. By summarizing the capability indices of each technology field, a multi-dimensional capability vector is constructed for each R&D personnel to describe their comprehensive technical capabilities in different technical directions, providing basic data for subsequent team-project matching analysis.

[0025] S2-3: Based on project task breakdown information, milestone data, and R&D activity records, calculate project progress indicators, represented as: PPI. p =∑ s=1 S w s *(T completed,s / T total,s ); among which, PPI p T represents the project progress indicators for project p; completed,s Indicates the amount of task completed in stage s; T total,s w represents the total amount of work in phase s.s The stage weights are S; S represents the total number of stages in project p. The aforementioned stage weighting indicators not only reflect the task completion rate, but can also be weighted and adjusted based on the quality of stage deliverables, thereby increasing the sensitivity to the actual implementation of the project.

[0026] The project progress indicators are used to reflect the degree of completion of tasks at each stage of the project.

[0027] S2-4: Normalize personnel capability indicators and project progress indicators: Unify indicators of different dimensions to a scale range of 0-1, forming a multi-dimensional evaluation vector V. p,s Characterized as: V p,s =[TAI i,t * PPI p ]; Among them, TAI i,t * This represents the average value of an index indicating an individual's competence in the technical field.

[0028] This multidimensional evaluation vector provides a comprehensive assessment of each project stage, offering a quantitative basis for subsequent team-project matching analysis, output prediction, and project progress optimization.

[0029] The indicator system can be dynamically updated. During project execution, new R&D behavior data (such as document submissions, code submissions, and experimental records) are collected, and personnel capability indices, output quality indicators, and project progress indicators are updated in real time, thereby forming a closed-loop data-driven R&D project management and analysis system.

[0030] S3. Based on the multidimensional evaluation index system, analyze the historical project participation data and achievement data of R&D personnel, calculate the capability feature vector of R&D personnel, and construct an R&D capability profile model. In S3, based on the aforementioned multidimensional evaluation index system, the historical project participation data and achievement data of R&D personnel are analyzed to calculate the R&D personnel capability feature vector and construct an R&D capability profile model. Specifically, this includes the following steps: S3-1: Based on S2, a multi-dimensional evaluation index system for personnel capability and project progress is constructed. This system analyzes the historical project data of R&D personnel, extracts the task contribution, role, quality of project stage results, and project completion status of each R&D personnel in each project, and classifies the projects according to the technical field, thereby obtaining a set of R&D personnel participation records in different technical fields. S3-2: Arrange the R&D personnel's capability indices across all technical fields according to the preset technical field set T={t1,t2,…,tm}, and construct the R&D personnel capability feature vector Fi: Fi=[TAI i,t1 TAI i,t2 ,…,TAI i,tm ]; where t1, t2, ..., tm represent the preset 1st, 2nd, and mth technical fields; TAI i,t1 TAI i,t2 ,…,TAI i,tm This represents the capability index of researcher i in the 1st, 2nd, and mth technical fields; The R&D personnel capability feature vector Fi is used to quantify the distribution of their capabilities in different technical fields.

[0031] S3-3: Based on the aforementioned R&D personnel capability feature vector, and integrating the historical output quantity and project execution performance of R&D personnel, according to the R&D personnel's project output quality index Q... p Project progress indicator PPI p Calculate the comprehensive ability score (CA) of R&D personnel. i : CA i =α*(Σ p∈Pi Q p ) / |P i |+β*(Σ p∈Pi PPI p ) / |P i |; Among them, P i This represents the set of projects in which R&D personnel i participates; α and β are weighting coefficients, satisfying α+β=1; this comprehensive ability score can reflect the overall performance of R&D personnel in terms of output capability and project execution capability.

[0032] S3-4: Combine the R&D personnel's capability feature vector Fi with the R&D personnel's comprehensive capability score CA i Combine the vectors to form a comprehensive capability representation vector Ai for R&D personnel: Ai=[Fi,CAi]; calculate the similarity between different comprehensive capability representation vectors of R&D personnel, perform capability clustering analysis on R&D personnel, and construct an R&D capability profile model; among them, the capability similarity between R&D personnel i1 and R&D personnel i2 is calculated by cosine similarity.

[0033] Specifically, R&D personnel are clustered based on the similarity to form a capability profile of R&D personnel that can represent the capability structure characteristics of different R&D personnel. The R&D capability profile model is stored in the science and technology management platform for subsequent team-project matching analysis, R&D resource optimization and allocation, and project outcome prediction analysis.

[0034] S4. Based on the R&D capability profile model and the technical requirements of the current project, calculate the matching relationship between R&D personnel and the project, obtain the team-project matching score, and generate R&D team configuration suggestions.

[0035] In S4, based on the R&D capability profile model and the technical requirements of the current project, the matching relationship between R&D personnel and the project is calculated to obtain a team-project matching score, and R&D team configuration suggestions are generated. Specifically, this includes the following steps: S4-1: Extract the technical requirements features of the target project for the R&D team to be configured; based on the project's technical solution description, task decomposition information, and project technical field tags, identify the set of technical fields Tp involved in the project. Tp = {t1, t2, ..., tm1}; t1, t2, tm1 represent the 1st, 2nd, and m1th technical fields involved in project p, respectively; Tp ⊆ T; Demand weights are assigned based on the importance of different technical fields in the project tasks, constructing a project technical demand vector. The technical demand vector Rp of project p is represented as: Rp = [w p,t1 ,w p,t2 ,…,w p,tm1 ];w p,t1 ,w p,t2 ,…,w p,tm1 Let w represent the demand weights of technical fields t1, t2, and tm1 in project p, respectively, satisfying: p,t1 +w p,t2 +…+w p,tm1 =1; S4-2: Using the R&D capability profile constructed in step S3, retrieve the capability indices for the 1st, 2nd, and m1th technical fields involved in project p, and extract the capability feature vectors Fi,p for each R&D personnel. Fi,p=[TAI i,t1 TAI i,t2 ,…,TAI i,tm1 By matching and analyzing the capability feature vector of R&D personnel with the technical requirement vector of the project, the capability matching degree r between R&D personnel i and project p is calculated: This is represented as: r=(Fi,p∙Rp) / (∥Fi,p∥∙∥Rp∥); where ∥Fi,p∥ and ∥Rp∥ represent the norms of the corresponding vectors, respectively. This indicator reflects the degree of similarity between the R&D personnel's skill set and the project's technical requirements.

[0036] S4-3: Based on the capability matching degree between R&D personnel i and project p and the team size required for the project, R&D personnel whose capability matching degree r is greater than a preset matching threshold are added to the candidates to generate the candidate R&D team G for project p.

[0037] Among these, as a preferred option, recommending the optimal R&D team for the N target projects currently requiring R&D team configuration includes the following steps: Step S1: Extract technical requirement features for each of the N target projects of the current R&D team to be configured; identify the technical fields involved in the project based on the project technical solution description, task decomposition information and project technical field tags; set requirement weights according to the importance of different technical fields in the project tasks; sort the N target projects in descending order based on the technical field with the highest requirement weight, and denote it as the target project sequence Y; Step S2: Based on the methods in S4-2 to S4-3, generate D candidate R&D teams for the target project; using the technical field of the target project sequence Y as the benchmark, sort the R&D personnel according to the capability index of the technical field from largest to smallest to obtain the R&D personnel capability feature sequence corresponding to the project sequence. The R&D personnel capability feature sequence of the D candidate R&D teams corresponding to the target project sequence Y is denoted as X=[X1,X2,…,Xd,…,XD], where Xk represents the R&D personnel capability feature sequence of the d-th candidate R&D team. Step S3: For any target project in the target project sequence Y involving h technical fields, each candidate R&D team shall include at least h R&D personnel. The capability index of R&D personnel in the same technical field shall be represented by the average value. The correlation ρ between the target project sequence and the capability characteristic sequence of R&D personnel of the candidate R&D team shall be calculated by the Pearson correlation coefficient method. The candidate R&D team with the largest |ρ| shall be recommended as the optimal R&D team.

[0038] It should be noted that sorting the N target projects in descending order based on the technology field with the highest demand weight can yield different target project sequences. In practice, a unique target project sequence can be obtained by setting key technology fields and sorting them in descending order by setting the capability index of the key technology fields. The Pearson correlation coefficient method is used in this application to calculate the correlation between the demand weight of the technical field and the capability index of R&D personnel. The two sequences must be of the same length, and the average value of multiple capability indices in the same technical field is taken. The Pearson correlation coefficient is [-1,1]. This application needs to ensure a positive correlation, so the candidate R&D team with the largest |ρ| will be recommended as the optimal R&D team.

[0039] For example, to address common industry problems such as misallocation of R&D resources, inefficient team configuration, and a disconnect between output and project needs, a technology company specializing in intelligent devices and big data management implemented an intelligent matching system for R&D personnel and projects. The specific implementation is as follows: The company leverages a unified identity authentication system to integrate three core data sources: project management, human resources, and technological achievement management. Under strict adherence to data security standards and matching user roles and permissions, authorized data collection is completed. For 12 ongoing research projects (covering three technical fields: intelligent robot control systems, public data management platforms, and intelligent building monitoring equipment), core project data such as project number, technical field, stage status, and task breakdown are collected. For 86 R&D personnel, basic information such as unique personnel identifiers, technical field tags, team affiliation, and historical project participation records are collected. Simultaneously, technical documents and achievement reports from each project stage are collected, and each piece of data is timestamped and has a source identifier to ensure data traceability.

[0040] By combining agile R&D with engineering implementation in enterprises, a multi-dimensional quantitative evaluation index is constructed. Regarding deliverable quality assessment, the project is divided into eight deliverable types based on four stages: project initiation, development, testing, and acceptance. These deliverables include technical documents, prototypes, test reports, and acceptance plans. A weighted aggregate score (Q) is calculated to determine the quality of deliverables at each stage. p The weight of core technological achievements (prototypes) is set at 0.25, and the weight of non-core auxiliary achievements (such as technical documents) is set at 0.05, so as to accurately quantify the technological value and innovation level of each project.

[0041] In the personnel capability quantification phase, the contribution and output quality scores of each R&D personnel in their respective technical fields are statistically analyzed, and the capability index (TAI) is calculated. i,t For example, a senior system development engineer has a capability index of 0.89 in the field of "public data management," while a smart device R&D engineer has a capability index of 0.76 in the field of "building intelligent monitoring," forming a personnel capability vector covering three major technology areas. Regarding project progress indicators, the PPI for each project is calculated based on task breakdown information and milestone data. p Simultaneously, the progress indicators are weighted and adjusted based on the quality of phased deliverables to avoid the bias of simply measuring project progress by task completion. Finally, personnel capability indicators and project progress indicators are normalized to the 0-1 range, generating a multi-dimensional evaluation vector V. p,s It also supports real-time updates during project execution, building a closed-loop data-driven system.

[0042] Based on the established multi-dimensional evaluation index system, we deeply analyzed the historical project data of R&D personnel to construct capability profiles. First, we extracted the participation records of 86 R&D personnel in various technical fields, arranged their capability indices according to a preset technical field set (intelligent control systems, public data management, and building intelligent monitoring), and generated a capability feature vector Fi for each person. Second, we integrated the quantity of historical outputs with project execution performance, setting a weight of α=0.65 for output quality and β=0.35 for project progress, and calculated the comprehensive capability score CA. iThis reflects the R&D personnel's output and project execution capabilities. Finally, the capability feature vector and the comprehensive score are combined into a comprehensive representation vector Ai. The cosine similarity is used to calculate the capability similarity between personnel, and cluster analysis is completed. Finally, five types of R&D capability profiles are divided into "core technology breakthrough type", "multi-domain all-round type", "engineering implementation type", "innovative R&D type" and "auxiliary support type". The profile model is stored in the R&D management platform to realize visual query and call.

[0043] For the four newly launched target projects (two intelligent robot control system projects, one public data management platform upgrade project, and one building intelligent monitoring equipment optimization project), precise matching was carried out. First, the technical requirements characteristics of each project were extracted to construct a technical requirements vector Rp: the weight of the "intelligent control system" requirement for the intelligent robot control system project was set to 0.7, and the weight of the "public data management" requirement was set to 0.3; the weight of the "public data management" requirement for the public data management platform upgrade project was set to 0.8, and the weight of the "intelligent control system" requirement was set to 0.2.

[0044] Subsequently, the R&D capability profiles are invoked to calculate the capability matching degree r between R&D personnel and projects (based on vector dot product and norm). Personnel with a matching degree higher than 0.75 are selected as candidates, generating four candidate teams. Finally, projects are sorted in descending order according to the technology field with the highest project requirement weight. The correlation ρ between the project sequence and the capability feature sequence of candidate teams is calculated using the Pearson correlation coefficient, prioritizing the team configuration with the highest correlation. For example, the intelligent robot control system project is matched with a combination of personnel profiles of "core technology breakthrough" and "multi-domain all-rounder," while the building intelligent monitoring equipment optimization project is matched with a combination of personnel profiles of "engineering implementation" and "intelligent control system." Ultimately, this achieves precise matching between teams and projects, effectively improving personnel allocation efficiency and project delivery quality.

[0045] A data intelligence analysis system for a technology management platform includes a data acquisition and processing module, an indicator construction module, a capability profile construction module, and a project matching analysis module; The data acquisition and processing module is used to acquire R&D-related data from multiple business systems, and to complete data preprocessing and correlation modeling, providing a unified data foundation for subsequent analysis; The indicator construction module is used to construct a multi-dimensional evaluation indicator system for personnel capabilities and project progress, so as to realize the quantitative expression of key indicators in the R&D process. The capability profile building module is used to model the capabilities of R&D personnel in multiple dimensions to form an R&D capability profile model; The project matching analysis module is used to perform matching analysis between R&D personnel and projects, and to generate candidate R&D teams.

[0046] Furthermore, it also includes a team optimization recommendation module; the team optimization recommendation module is used to recommend the optimal R&D team based on correlation analysis in multi-project scenarios.

[0047] The correlation analysis was performed using the Pearson correlation coefficient method.

[0048] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A data intelligent analysis method for a technology management platform, characterized in that: Specifically, the steps include the following: S1. Collect project data, R&D personnel data, and technological achievement data from the R&D management platform of technology companies, clean, deduplicate, and standardize them, and establish the relationship data between R&D personnel and projects based on personnel participation relationships; S2. Based on the project data, R&D personnel data, and technological achievement data, construct a multi-dimensional evaluation index system of personnel capability and project progress. The personnel capability index is used to characterize the technical capability level of R&D personnel in different technical fields, and the project progress index is used to characterize the degree of task completion of the R&D project at different stages. S3. Based on the multidimensional evaluation index system, analyze the historical project participation data and achievement data of R&D personnel, calculate the capability feature vector of R&D personnel, and construct an R&D capability profile model. S4. Based on the R&D capability profile model and the technical requirements of the current project, calculate the matching relationship between R&D personnel and the project, obtain the team-project matching score, and generate R&D team configuration suggestions.

2. The data intelligent analysis method for a technology management platform according to claim 1, characterized in that: In S1, project data, R&D personnel data, and technological achievement data are collected from the R&D management platform of technology companies. This data is then cleaned, deduplicated, and standardized. Furthermore, data linking R&D personnel to projects is established based on their involvement. Specifically: Data related to R&D activities is collected from various data source systems. Project data includes at least project number, project name, technical field, project stage status, and task breakdown information. Data quality processing is performed on the collected data, including detecting and completing missing fields and marking anomalies, identifying duplicate data through key field comparison and retaining the latest data version, and detecting outliers in fields such as project cycle and number of deliverables. After data cleaning is completed, data from different systems are processed in a unified and standardized manner. Establish the relationship data between R&D personnel and projects based on the records of R&D personnel's participation in projects, and record the personnel's participation role, participation time and task contribution information in the relationship data. Construct a triple of participation role, participation time and task contribution information as the relationship data.

3. The data intelligent analysis method for a technology management platform according to claim 2, characterized in that: include: The system identifies data source systems within the enterprise that are involved in R&D management, obtains data access tokens through the enterprise's unified identity authentication mechanism, matches user role permissions based on the access tokens, generates corresponding data access control policies, and authorizes access to relevant data in the project management system, human resources system, and technology achievement management system while ensuring data security.

4. The data intelligent analysis method for a technology management platform according to claim 3, characterized in that: In S2, based on the project data, R&D personnel data, and technological achievement data, a multi-dimensional evaluation index system of personnel capability and project progress is constructed. The personnel capability index characterizes the technical capability level of R&D personnel in different technical fields, and the project progress index characterizes the degree of task completion at different stages of the R&D project. Specifically, this includes the following steps: S2-1: Based on the project's phase status, construct a project deliverable quality assessment model, specifically characterized as follows: Q p =∑ k=1 ns w k *v k ; where Q p This represents the quality score of the R&D personnel for each stage of the project; s represents the stage of the project; ns represents the total number of deliverable types included in stage s; k represents the type of deliverables in the stage; v k Scoring of interim results; k Weights for different types of stage deliverables; S2-2: Based on collected project data, R&D personnel data, and technological achievement data, quantitatively evaluate the capabilities of R&D personnel in different technical fields to form personnel capability indicators; the specific method is as follows: statistically analyze the task contribution and stage achievement quality score of each R&D personnel in the project set they participate in, and calculate the personnel's capability index (TAI) in the technical field t. i,t Where i represents R&D personnel; p represents projects; C i,p This indicates the researcher's contribution to project p; P i t Represents the set of items in the technical field t; S2-3: Based on project task breakdown information, milestone data, and R&D activity records, calculate project progress indicators, represented as: PPI. p =∑ s=1 S w s *(T completed,s / T total,s ); among which, PPI p T represents the project progress indicators for project p; completed,s Indicates the amount of task completed in stage s; T total,s w represents the total amount of work in phase s. s The stage weights are S; S represents the total number of stages in project p. S2-4: Unify the different dimensional indicators to a scale range of 0-1, forming a multidimensional evaluation vector V. p,s Characterized as: V p,s =[TAI i,t * PPI p ]; among them, TAI i,t * This represents the average value of an index indicating an individual's competence in the technical field.

5. The data intelligent analysis method for a technology management platform according to claim 4, characterized in that: In S3, based on the aforementioned multidimensional evaluation index system, the historical project participation data and achievement data of R&D personnel are analyzed to calculate the R&D personnel capability feature vector and construct an R&D capability profile model. Specifically, this includes the following steps: S3-1: Based on S2, a multi-dimensional evaluation index system for personnel capability and project progress is constructed. This system analyzes the historical project data of R&D personnel, extracts the task contribution, role, quality of project stage results, and project completion status of each R&D personnel in each project, and classifies the projects according to the technical field, thereby obtaining a set of R&D personnel participation records in different technical fields. S3-2: Arrange the R&D personnel's capability indices across all technical fields according to the preset technical field set T={t1,t2,…,tm}, and construct the R&D personnel capability feature vector Fi: Fi=[TAI i,t1 TAI i,t2 ,…,TAI i,tm ]; where t1, t2, ..., tm represent the preset 1st, 2nd, and mth technical fields; TAI i,t1 TAI i,t2 ,…,TAI i,tm This represents the capability index of researcher i in the 1st, 2nd, and mth technical fields; S3-3: Based on the aforementioned R&D personnel capability feature vector, and according to the Q-index of the results of the R&D personnel's participation in the project. p Project progress indicator PPI p Calculate the comprehensive ability score (CA) of R&D personnel. i : CA i =α*(Σ p∈Pi Q p ) / |P i |+β*(Σ p∈Pi PPI p ) / |P i |;wherein, P i Let represent the set of projects in which R&D personnel i participates; α and β are weighting coefficients, satisfying α+β=1; S3-4: Combine the R&D personnel's capability feature vector Fi with the R&D personnel's comprehensive capability score CA i The vectors are combined to form a comprehensive capability representation vector Ai: Ai=[Fi,CAi]; the similarity between different comprehensive capability representation vectors of R&D personnel is calculated by cosine similarity, and capability clustering analysis is performed on R&D personnel to construct an R&D capability profile model.

6. The data intelligent analysis method for a technology management platform according to claim 5, characterized in that: In S4, based on the R&D capability profile model and the technical requirements of the current project, the matching relationship between R&D personnel and the project is calculated to obtain a team-project matching score, and R&D team configuration suggestions are generated. Specifically, this includes the following steps: S4-1: Extract the technical requirements features of the target project for the R&D team to be configured; based on the project's technical solution description, task decomposition information, and project technical field tags, identify the set of technical fields Tp involved in the project. Tp={t1,t2,…,tm1}; t1, t2, tm1 represent the first, second, and m1 technical fields involved in project p, respectively; Tp⊆T; Based on the importance of different technical fields in the project tasks, requirement weights are set to construct the project's technical requirement vector, where the technical requirement vector Rp of project p is represented as: Rp=[w p,t1 ,w p,t2 ,…,w p,tm1 ];w p,t1 ,w p,t2 ,…,w p,tm1 Let w represent the demand weights of technical fields t1, t2, and tm1 in project p, respectively, satisfying: p,t1 +w p,t2 +…+w p,tm1 =1; S4-2: Using the R&D capability profile constructed in step S3, retrieve the capability indices for the 1st, 2nd, and m1th technical fields involved in project p, and extract the capability feature vectors Fi,p for each R&D personnel. Fi,p=[TAI i,t1 TAI i,t2 ,…,TAI i,tm1 By matching and analyzing the capability feature vector of R&D personnel with the technical requirement vector of the project, the capability matching degree r between R&D personnel i and project p is calculated: This is represented as: r=(Fi,p∙Rp) / (∥Fi,p∥∙∥Rp∥); where ∥Fi,p∥ and ∥Rp∥ represent the norms of the corresponding vectors, respectively. S4-3: Based on the capability matching degree between R&D personnel i and project p and the team size required for the project, R&D personnel whose capability matching degree r is greater than a preset matching threshold are added to the candidates to generate the candidate R&D team G for project p.

7. The data intelligent analysis method for a technology management platform according to claim 6, characterized in that: Also includes: Recommend the optimal R&D team for N target projects that currently require R&D team allocation, specifically: Step S1: Extract the technical requirements features for each of the N target projects of the current R&D team to be configured; identify the technical fields involved in the project based on the project technical solution description, task decomposition information and project technical field tags; Demand weights are assigned based on the importance of different technical fields in the project tasks; N target projects are sorted in descending order of their demand weights based on the technical field with the highest demand weight, and this sequence is denoted as target project sequence Y. Step S2: Based on the methods in S4-2 to S4-3, generate D candidate R&D teams for the target project; using the technical field of the target project sequence Y as the benchmark, sort the R&D personnel according to the capability index of the technical field from largest to smallest to obtain the R&D personnel capability feature sequence corresponding to the project sequence. The R&D personnel capability feature sequence of the D candidate R&D teams corresponding to the target project sequence Y is denoted as X=[X1,X2,…,Xd,…,XD], where Xk represents the R&D personnel capability feature sequence of the d-th candidate R&D team. Step S3: For any target project in the target project sequence Y involving h technical fields, each candidate R&D team shall include at least h R&D personnel. The capability index of R&D personnel in the same technical field shall be represented by the average value. The correlation ρ between the target project sequence and the capability characteristic sequence of R&D personnel of the candidate R&D team shall be calculated by the Pearson correlation coefficient method. The candidate R&D team with the largest |ρ| shall be recommended as the optimal R&D team.

8. A data intelligent analysis system for a technology management platform, employing the data intelligent analysis method for a technology management platform as described in any one of claims 1-7, characterized in that: It includes a data acquisition and processing module, an indicator construction module, a capability profile construction module, and a project matching analysis module; The data acquisition and processing module is used to acquire R&D-related data from multiple business systems, and to complete data preprocessing and correlation modeling, providing a unified data foundation for subsequent analysis; The indicator construction module is used to construct a multi-dimensional evaluation indicator system for personnel capabilities and project progress, so as to realize the quantitative expression of key indicators in the R&D process. The capability profile building module is used to model the capabilities of R&D personnel in multiple dimensions to form an R&D capability profile model; The project matching analysis module is used to perform matching analysis between R&D personnel and projects, and to generate candidate R&D teams.

9. The data intelligent analysis system for a technology management platform according to claim 8, characterized in that: It also includes a team optimization recommendation module; the team optimization recommendation module is used to recommend the optimal R&D team based on correlation analysis in multi-project scenarios.

10. The data intelligent analysis system for a technology management platform according to claim 9, characterized in that: The correlation analysis was performed using the Pearson correlation coefficient method.

Images