Search

CN-121998460-A - Urban talent development evaluation method, equipment and storage medium

CN121998460ACN 121998460 ACN121998460 ACN 121998460ACN-121998460-A

Abstract

The application relates to a city talent development evaluation method, equipment and a storage medium, wherein data of three-level indexes in each index system are collected and acquired by calling a constructed city talent development evaluation index system; the method comprises the steps of obtaining data of three-level indexes in each index system according to data distribution of the three-level indexes in each index system, carrying out standardization processing on the data to obtain standardized values of the three-level indexes in each index system, determining weights of the three-level indexes in each index system by adopting a factor analysis method, and determining urban talent development evaluation scores according to the standardized values and the weights of the three-level indexes in each index system. The application realizes the multi-level, multi-dimensional, quantitative and systematic comprehensive evaluation of each innovation main body and innovation development environment of the city by constructing the talent development evaluation system of the city. The index weights of all levels are determined by adopting a factor analysis method, so that the analysis and calculation process is simplified, the interference of subjective judgment of an expert is reduced, and the evaluation result is more objective and accurate.

Inventors

  • Zheng Zhehui
  • QIU YU
  • LIU DEBING

Assignees

  • 北京智谱华章科技股份有限公司

Dates

Publication Date
20260508
Application Date
20241031

Claims (10)

  1. 1. The city talent development evaluation method is characterized by comprising the following steps of: Invoking a constructed urban talent development evaluation index system, wherein the urban talent development evaluation index system comprises a talent quality index system, a mechanism platform index system and an urban development index system, and each layer of index system comprises a first-level index, a second-level index and a third-level index; Collecting and acquiring data of three-level indexes in each index system; according to the data distribution of the three-level indexes in each index system, carrying out standardization processing on the data to obtain the standardized numerical values of the three-level indexes in each index system; Determining the weight of three-level indexes in each index system by adopting a factor analysis method, wherein a factor score is calculated by adopting a variable highly related to a specific factor in the factor analysis method; determining urban talent development evaluation scores according to standardized values and weights of three-level indexes in each index system: Wherein pi represents the urban talent development evaluation score, gamma f represents the weight of the three-level index in each index system, and M f represents the standardized numerical value of the three-level index in each index system.
  2. 2. The urban talent development evaluation method according to claim 1, wherein said determining weights of three-level indexes in each index system by a factor analysis method comprises: establishing a factor analysis model of three-level indexes of each city hierarchy: Wherein Y i represents an ith three-level index in each hierarchy, X j represents a factor with a jth feature root greater than 1, epsilon i represents an error vector, and alpha ij represents the load of an ith variable on the jth factor; The factor analysis model is expressed in matrix form as Y=AX+epsilon Wherein, the Is a three-level index vector which is used for the index of the, As a matrix of the factor load, As a vector of the factors, Error vector; extracting common factors by a principal component factor method, solving a correlation matrix, and calculating characteristic roots and contribution rates: Wherein, the Vectors which are formed by arranging the extracted main components from large to small according to characteristic roots, A component of the feature vector corresponding to the feature root lambda i (λ 1 …λ i > 0) of the correlation matrix composed of the original index Y; The number w of common factors is determined according to the characteristic root selection of the correlation coefficient matrix (lambda w is more than or equal to 1), and the variables are classified and named according to the load of each factor on each variable, namely: Y 1 …Y i ∈M 1 Wherein M 1 represents a factor 1, Y 1 …Y i is a variable having the highest factor loading coefficient on the factor 1; calculating the variance contribution rate and the accumulated variance interpretation rate of the principal components, wherein the variance contribution rate of the w-th principal component is as follows: the cumulative variance interpretation rate is: Factor rotation is carried out on the variable by using a maximum variance orthogonal rotation method, so as to obtain an interpretable common factor and a factor load coefficient matrix A' =A×T after rotation, wherein T is an orthogonal matrix; calculating index weights and factor scores by regression estimation according to the factor load coefficient matrix A' after rotation: Wherein, the For the factor score vector obtained by regression estimation, For the three-level index weight matrix, Three-level index vectors belonging to the corresponding factor class; calculating the weight of each factor according to the variance contribution rate and the accumulated variance contribution rate of each common factor after rotation:
  3. 3. The talent development evaluation method of city according to claim 1, further comprising determining scores of each city in each dimension according to standardized values and weights of three-level indexes in each index system, such as talent attraction, development platform, talent cultivation, occupation promotion, life guarantee and reasonable configuration; The scores of each city in each dimension of talent attraction, development platform, talent cultivation, occupation promotion, life guarantee and reasonable configuration are calculated by adopting the following steps: or pi d ＝β 1d Y 1d +β 2d Y 2d +…+β id Y id Wherein pi d represents the score of the city in the dimension d, gamma f represents the weight of the factor corresponding to the dimension d, M f represents the score of the factor corresponding to the dimension d, beta id represents the weight of the three-level index contained in the dimension d, and Y id represents the standardized value of the three-level index contained in the dimension d.
  4. 4. The method for evaluating the talent development of cities according to claim 1, wherein said collecting and acquiring data of three levels of indicators in each indicator system comprises: Web crawlers based on natural language processing technology and text analysis large models perform structured extraction on webpage data to obtain data of three-level indexes.
  5. 5. The urban talent development evaluation method according to claim 4, wherein the web crawler based on the natural language processing technology and the text segmentation model performs structural extraction on the web page data, and the obtaining of the data of the three-level index comprises: identifying abnormal data in the webpage data by using a natural language processing technology, and processing the abnormal data; converting the data from different sources and formats into a standard format suitable for analysis and calculation; and extracting key features in the corresponding data of each city by utilizing the context understanding capability of the deep learning large model, and carrying out data enhancement.
  6. 6. The urban talent development evaluation method according to any one of claims 1 to 5, wherein said normalizing the data according to the data distribution of the three-level index in each index system, obtaining the normalized value of the three-level index in each index system comprises: Carrying out standardized treatment on the data by adopting a Max-Min dimensionless method to obtain standardized values of three-level indexes in each index system: wherein Z represents the standardized value of each three-level index, x is the initial value of the corresponding three-level index, and min and max respectively correspond to the lower limit and the upper limit of the index distribution interval.
  7. 7. The urban talent development evaluation method according to any one of claims 1 to 5, wherein the secondary index in the talent quality index system includes a talent achievement index, a talent discipline index, a talent interdisciplinary index, a talent society index, a talent cooperation index, and a talent international index; In the mechanism platform index system, the secondary index comprises a mechanism platform talent scale index, a mechanism platform scientific research index, a mechanism platform social index, a mechanism platform cooperation index and a mechanism platform international index; In the urban development index system, the secondary indexes comprise an urban talent attraction index, an urban development platform index, an urban talent culture index, an urban occupation promotion index, an urban life guarantee index and an urban reasonable configuration index.
  8. 8. The method for evaluating the talent development of cities according to claim 7, wherein said collecting and acquiring data of three levels of indicators in each indicator system comprises: in a talent quality index system, collecting data of an H index, liveness, patent quantity, ESI high cited paper percentage, patent valuation, average percentile, cow cited ratio, first author percentage, cooperation efficiency index, obstetrical research cooperation percentage, journal standardization quotation influence, discipline diversity, discipline standardization quotation influence, field authority prize number, network search quantity, high-level talent index and international cooperation percentage; Collecting data of academic talent scale, commercial talent rule quantity, scientific literature publishing quantity, scientific research personnel quantity, patent quantity, academic influence of science relative to global level, first author percentage, cooperation index, obstetric and academic cooperation percentage, number of winnings, international cooperation percentage and cooperation country number in an institution platform index system; In an urban development index system, collecting data of research personnel quantity, research personnel acceleration, high-level talent quantity, excellent talent quantity, backup talent quantity, scientific literature publishing quantity, ESI high-lead paper percentage, patent evaluation, scientific research institution research post, world 500 strong enterprises, chinese civil enterprises 500 strong enterprises, national key laboratory quantity, average R & D cost, basic research cost input proportion, national achievement conversion platform quantity, average talent ownership, talent title promotion time, average talent wages, regional production total value acceleration, average talent possession bed number, average talent commute time, average residence area, population density, average talent park green area, air quality index annual fine rate, average cultural activity place quantity, middle and primary school students ratio, high school teacher ratio, average education year, education cost input proportion, average talent proportion, population retention rate, lead index and flow proportion.
  9. 9. A talent development evaluation device comprising a memory and a processor, said memory storing a computer program which, when executed by said processor, implements a talent development evaluation method according to any one of claims 1-8.
  10. 10. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, implements the urban talent development evaluation method according to any one of claims 1-8.

Description

Urban talent development evaluation method, equipment and storage medium Technical Field The application belongs to the technical field of urban development, and particularly relates to an urban talent development evaluation method, equipment and a storage medium. Background The existing talent evaluation system adopts a single index method, for example, the number of published papers is used as a standard for evaluating talents, the talent scale is used as a standard for evaluating urban talent competitiveness, index weights are determined based on a Delphi method so as to construct evaluation indexes, the index weights are determined by consensus through multiple rounds of expert opinion communication and feedback, the pack overtaking weights are determined based on an entropy method so as to construct evaluation indexes, and the index weights are determined according to the information quantity provided by each index. However, the prior art method has the following drawbacks: The single index method is used for evaluating talent quality or urban talent development condition only by using a single index, the evaluation result is too complete, the evaluation results obtained by using different indexes are not comparable, the evaluation results are easy to operate, the flexibility is lacking, and the requirements of different fields on talents or cities are different, so that the evaluation results cannot be compared transversely. The Delphi method generally requires multiple rounds of expert anonymous investigation and feedback, takes long time, can be influenced by subjectivity in the process of selecting the expert and determining the weight by the expert, and has poor interpretation for non-expert groups based on expert consensus as a result. The entropy method is that the weight result is inconsistent with the actual importance of the indexes due to complete dependence on the data, and the weight distribution is unreasonable due to the fact that the indexes are mutually independent and the correlation among the indexes is ignored. Therefore, how to realize the multi-dimensional, quantitative and systematic comprehensive evaluation of urban talent development is a subject worthy of research in the existing urban planning development. Disclosure of Invention In view of the above analysis, the embodiment of the invention aims to provide an urban talent development evaluation method, and aims to solve the defects of single dimension, strong subjectivity and poor evaluation effect of the existing talent evaluation method. The first aspect of the application provides a city talent development evaluation method, which comprises the following steps: Invoking a constructed urban talent development evaluation index system, wherein the urban talent development evaluation index system comprises a talent quality index system, a mechanism platform index system and an urban development index system, and each layer of index system comprises a first-level index, a second-level index and a third-level index; Collecting and acquiring data of three-level indexes in each index system; according to the data distribution of the three-level indexes in each index system, carrying out standardization processing on the data to obtain the standardized numerical values of the three-level indexes in each index system; Determining the weight of three-level indexes in each index system by adopting a factor analysis method, wherein a factor score is calculated by adopting a variable highly related to a specific factor in the factor analysis method; determining urban talent development evaluation scores according to standardized values and weights of three-level indexes in each index system: Wherein pi represents the urban talent development evaluation score, gamma f represents the weight of the three-level index in each index system, and M f represents the standardized numerical value of the three-level index in each index system. Optionally, determining the weights of the three-level indexes in each index system by using a factor analysis method includes: establishing a factor analysis model of three-level indexes of each city hierarchy: Wherein Y i represents an ith three-level index in each hierarchy, X j represents a factor with a jth feature root greater than 1, epsilon i represents an error vector, and alpha ij represents the load of an ith variable on the jth factor; The factor analysis model is expressed in matrix form as Y=AX+epsilon Wherein, the Is a three-level index vector which is used for the index of the,As a matrix of the factor load,As a vector of the factors,Error vector; extracting common factors by a principal component factor method, solving a correlation matrix, and calculating characteristic roots and contribution rates: Wherein, the Vectors which are formed by arranging the extracted main components from large to small according to characteristic roots,A component of the feature vector corresponding to the fea