CN-121998601-A - Personnel post matching method, device, equipment and medium

CN121998601ACN 121998601 ACN121998601 ACN 121998601ACN-121998601-A

Abstract

The invention relates to the technical field of data analysis, and discloses a personnel post matching method, a device, equipment and a medium, wherein multimode resume data of a plurality of target candidates and post requirement data of a plurality of posts are obtained, and a post requirement weight matrix is constructed according to the post requirement data; converting the multi-mode resume data into candidate feature vectors, converting the post requirement data into post feature vectors, carrying out weighted fusion on the candidate feature vectors according to the post requirement weight matrix to obtain candidate weighted vectors, calculating the matching degree of the candidate weighted vectors and the post feature vectors, carrying out association relation feature extraction on the multi-mode resume data to obtain data association features, and generating target personnel post matching results according to the data association features and the matching degree. According to the invention, the data association characteristics are extracted, so that the limitation that only a single attribute is concerned in the traditional characteristic extraction scheme is broken through, and meanwhile, the matching accuracy of personnel and posts is improved.

Inventors

SUN CHENGCUI
CHEN ZHISHENG
LIU YI
CHEN XIANLI
YE YINGQI

Assignees

招商局金融科技有限公司

Dates

Publication Date: 20260508
Application Date: 20251224

Claims (10)

1. A person post matching method, comprising: acquiring multimode resume data of a plurality of target candidates and post requirement data of a plurality of posts, and constructing a post requirement weight matrix according to the post requirement data; Converting the multi-mode resume data into candidate feature vectors, converting the post requirement data into post feature vectors, and carrying out weighted fusion on the candidate feature vectors according to the post requirement weight matrix to obtain candidate weighted vectors; calculating the matching degree of the candidate weighting vector and the post feature vector; Carrying out association relation feature extraction on the multi-mode resume data to obtain data association features; and generating a target personnel post matching result according to the data association characteristic and the matching degree.
2. The person post matching method of claim 1, wherein said converting the multimodal resume data into candidate feature vectors comprises: Carrying out standardization processing on the multi-mode resume data to obtain standardized resume data, and extracting mode type characteristics corresponding to the standardized resume data; Converting text data in the standardized resume data into modal high-dimensional semantic vectors; And constructing a type feature space according to the modal type features, and carrying out semantic space mapping on the modal high-dimensional semantic vector according to the type feature space to obtain a candidate feature vector.
3. The personnel post matching method of claim 1, wherein the constructing a post demand weight matrix from the post demand data comprises: Extracting post scene description data in the post requirement data, and extracting scene characteristics of the post scene description data to obtain post scene characteristics; Screening candidate optional job evaluation dimensions in the post scene characteristics, and distributing weights for the candidate optional job evaluation dimensions to obtain a basic weight matrix; And carrying out weight adjustment on the basic weight matrix according to the post scene characteristics to obtain a post demand weight matrix.
4. The personnel post matching method as claimed in claim 1, wherein the step of performing weighted fusion on the candidate feature vector according to the post demand weight matrix to obtain a candidate weighted vector comprises the steps of: establishing a dimension mapping relation between the post demand weight matrix and the candidate feature vector; Carrying out importance weighting treatment on the candidate feature vector according to the dimension mapping relation to obtain a weighted candidate feature vector; and carrying out normalization processing on the weighted candidate feature vector to obtain a candidate weighted vector.
5. The person post matching method of claim 1, wherein said calculating a degree of matching of the candidate weight vector to the post feature vector comprises: Respectively carrying out depth semantic coding on the candidate weighting vector and the post feature vector to obtain a candidate coding vector and a post coding vector; Calculating the similarity of the candidate code vector and the post code vector; and carrying out normalization processing on the similarity, and converting the normalized similarity into matching degree.
6. The personnel post matching method of claim 1, wherein the extracting the association relation feature of the multi-mode resume data to obtain the data association feature comprises: Extracting candidate tenure basic information of the multi-mode resume data, and extracting post requirement information in the post requirement data; constructing a multi-type node set according to the candidate tenure basic information and the post demand information; Determining a skill association edge set and a company association edge set according to the multi-type node set; Constructing an abnormal composition according to the multi-type node set, the skill association side set and the company association side set; carrying out hierarchical architecture division on the heterogeneous graph, and constructing a topological relation map according to the divided architecture; converting each node in the topological relation map into an initial feature vector to obtain a node initial feature vector set; Performing high-order association feature fusion on each node in the node initial feature vector set to obtain an association fusion vector of each node; And splicing the association fusion vector into data association features.
7. The person post matching method of claim 1, wherein the generating a target person post matching result from the data correlation feature and the degree of matching comprises: splicing the candidate code vector and the post code vector corresponding to the matching degree into a fusion vector; Carrying out the same vector space projection on the fusion vector according to the data association characteristic to obtain a projection characteristic vector; performing attention fusion on the data association feature and the projection feature vector to obtain a comprehensive feature; Acquiring multiple target matching dimensions, and distributing weights for each target dimension in the multiple target matching dimensions to obtain target dimension weights of each target dimension; Weighting and fusing the comprehensive characteristics and the target dimension weights to obtain a multi-target comprehensive matching value; determining an initial personnel post matching scheme according to the multi-target comprehensive matching value and a preset target matching threshold value; Updating the multi-target comprehensive matching value according to a preset historical candidate target dimension weight to obtain an updated multi-target comprehensive matching value; and adjusting the initial personnel post matching scheme according to the updated multi-target comprehensive matching value to obtain a target personnel post matching result.
8. A person post matching device, comprising: the matrix construction module is used for acquiring multi-mode resume data of a plurality of target candidates and post requirement data of a plurality of posts, and constructing a post requirement weight matrix according to the post requirement data; The transformation fusion module is used for transforming the multi-mode resume data into candidate feature vectors, transforming the post requirement data into post feature vectors, and carrying out weighted fusion on the candidate feature vectors according to the post requirement weight matrix to obtain candidate weighting vectors; the matching degree calculation module is used for calculating the matching degree of the candidate weighting vector and the post feature vector; the data association feature extraction module is used for extracting association relation features of the multi-mode resume data to obtain data association features; And the personnel position result generating module is used for generating a target personnel position matching result according to the data association characteristics and the matching degree.
9. Computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the person position matching method according to any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the person position matching method of any of claims 1 to 7.

Description

Personnel post matching method, device, equipment and medium Technical Field The present invention relates to the field of data analysis technologies, and in particular, to a method, an apparatus, a device, and a medium for personnel post matching. Background Along with the continuous promotion of economic globalization and the continuous deepening of industrial structure adjustment, the demands of enterprises for talents are increasingly diversified and specialized, so that efficient and accurate personnel post matching is realized, and the requirements of enterprises and job seekers in the field of human resource management are met. In the existing personnel post matching scheme, manual screening and simple keyword matching modes are adopted, the recruiter needs to check a large number of resume one by one, the matching degree of the candidate and the post is judged manually, the mode is time-consuming and labor-consuming, personal subjective deviation is difficult to avoid, objectivity and consistency of an evaluation result cannot be guaranteed, the matching degree is judged by matching keywords in the resume with keywords in post requirements, and the matching mode cannot deeply understand the real capability of the candidate and the deep requirements of the post, so that the accuracy of the matching result is insufficient. Therefore, in order to meet the increasing demands of personnel post matching, the current personnel post matching method needs to be improved so as to solve the problems of low efficiency and insufficient accuracy of matching results of the existing method. Disclosure of Invention The invention provides a personnel post matching method, a device, equipment and a medium, which mainly solve the problem that the matching accuracy of personnel and posts is improved while breaking through the limitation that the traditional characteristics pay attention to single attribute. In a first aspect, a person post matching method is provided, including: acquiring multimode resume data of a plurality of target candidates and post requirement data of a plurality of posts, and constructing a post requirement weight matrix according to the post requirement data; Converting the multi-mode resume data into candidate feature vectors, converting the post requirement data into post feature vectors, and carrying out weighted fusion on the candidate feature vectors according to the post requirement weight matrix to obtain candidate weighted vectors; calculating the matching degree of the candidate weighting vector and the post feature vector; Carrying out association relation feature extraction on the multi-mode resume data to obtain data association features; and generating a target personnel post matching result according to the data association characteristic and the matching degree. In a second aspect, a person post matching device is provided, including: the matrix construction module is used for acquiring multi-mode resume data of a plurality of target candidates and post requirement data of a plurality of posts, and constructing a post requirement weight matrix according to the post requirement data; The transformation fusion module is used for transforming the multi-mode resume data into candidate feature vectors, transforming the post requirement data into post feature vectors, and carrying out weighted fusion on the candidate feature vectors according to the post requirement weight matrix to obtain candidate weighting vectors; the matching degree calculation module is used for calculating the matching degree of the candidate weighting vector and the post feature vector; the data association feature extraction module is used for extracting association relation features of the multi-mode resume data to obtain data association features; And the personnel position result generating module is used for generating a target personnel position matching result according to the data association characteristics and the matching degree. In a third aspect, a computer device is provided comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of a person position matching method as described above when executing the computer program. In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of a person position matching method as described above. According to the scheme realized by the personnel post matching method, the device, the computer equipment and the storage medium, the multi-mode resume data and post requirement data are obtained, the multi-dimensional information of the candidate and the post complete requirement are covered, so that matching deviation caused by single data dimension is avoided, the post requirement weight matrix is constructed, the dimension focused by post can be