CN-121998337-A - Scientific research talent and enterprise information matching method and system

CN121998337ACN 121998337 ACN121998337 ACN 121998337ACN-121998337-A

Abstract

The invention relates to the technical field of information technology and discloses a method and a system for matching talents of scientific researches with enterprise information, wherein the method comprises the steps of obtaining enterprise demand information and constructing an enterprise demand feature set; the method comprises the steps of obtaining scientific research achievement information and constructing an achievement feature set to generate an application scene prediction result, obtaining enterprise historical technical information and constructing an enterprise technical layout feature set to identify potential technical demand directions, obtaining scientific research talent information and constructing a scientific research talent capability feature set, mapping the feature set to a unified feature space to form an enterprise demand vector, an enterprise technical layout vector and a scientific research talent capability vector, carrying out matching analysis on the scientific research talent capability vector, the enterprise demand vector and the enterprise technical layout vector based on the unified feature space to generate a matching result set, and outputting matched scientific research talent information and associated scientific research achievement recommendation or cooperation butt joint information. The invention can improve the industrialization efficiency of the scientific research results.

Inventors

HUANG SHANSHI
SUN HAIXIANG
YANG JUAN
GUAN HONGCAI
ZHANG JIAHUI
ZHENG XIAOZHOU

Assignees

温州市工科院技术经纪服务有限公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. The method for matching the talents of scientific research with the enterprise information is characterized by comprising the following steps: Acquiring enterprise demand information of a target enterprise, wherein the enterprise demand information at least comprises technical field information, industry direction information, application scene description information and technical maturity requirement information, carrying out semantic analysis and structuring processing on the enterprise demand information, and constructing an enterprise demand feature set; Acquiring scientific research result information, wherein the scientific research result information at least comprises result technology profile information, technical field information of a result, technical maturity evaluation information and technical development trend information, analyzing the scientific research result information, constructing a result feature set, and generating an application scene prediction result based on the result feature set; acquiring historical technical information of a target enterprise, wherein the historical technical information at least comprises laid technical keyword information, existing technical direction information and published scientific research achievements or patent information of the enterprise, analyzing the historical technical information, constructing an enterprise technical layout feature set, and identifying potential technical demand directions based on the enterprise technical layout feature set; acquiring scientific research talent information, wherein the scientific research talent information at least comprises scientific research direction information, scientific research achievements or patent information and scientific research experience information, analyzing the scientific research talent information, and constructing a scientific research talent capability feature set; Mapping the enterprise demand feature set, the achievement feature set, the enterprise technical layout feature set and the scientific research person capacity feature set to a unified feature space to form an enterprise demand vector, an enterprise technical layout vector and a scientific research person capacity vector; Based on the unified feature space, carrying out matching analysis on the talent capability vector of the scientific research person and the enterprise demand vector and the enterprise technical layout vector to generate a matching result set between the scientific research person and a target enterprise; And outputting the talent information of the scientific research matched with the target enterprise based on the matching result set, and generating the recommendation information of the scientific research achievements or the cooperation docking information associated with the talent information of the scientific research.
2. The method for matching talents to enterprise information of claim 1, wherein constructing the enterprise demand feature set comprises: The enterprise demand information is disassembled into a demand segment set, wherein the demand segment set at least comprises a technical object segment, an application scene segment, a constraint condition segment and a maturity segment; Generating a segment evidence item aiming at each demand segment in the demand segment set, wherein the segment evidence item at least comprises a position index of the segment in enterprise demand information, a trigger word type of the segment and a context boundary of the segment; Constructing an enterprise demand evidence chain based on the segment evidence items, connecting the technical object segments, the application scene segments and the constraint condition segments according to the position index sequence by the enterprise demand evidence chain, and recording the connection types between the adjacent segments, wherein the connection types at least comprise causal connection, parallel connection and conditional connection; Under the condition that the same technical object segment corresponds to a plurality of different application scene segments in the enterprise demand evidence chain, splitting the enterprise demand evidence chain into a plurality of demand sub-chains, and respectively forming a group of candidate demand feature groups in the enterprise demand feature set by each demand sub-chain.
3. The method for matching talents to enterprise information according to claim 2, wherein determining technical field information and industry direction information in enterprise demand information comprises: extracting technical object core phrases based on technical object fragments in each requirement sub-chain, and extracting scene object core phrases based on application scene fragments; Performing field mapping on the technical object core phrase and the scene object core phrase respectively to obtain a technical field candidate set and an industry direction candidate set; under the condition that a plurality of candidates exist in the technical field candidate set and a unique main candidate cannot be formed, invoking a connection type of a requirement sub-chain to perform disambiguation judgment, wherein the disambiguation judgment comprises the steps of preferentially selecting the candidate consistent with the technical object core phrase mapping when the connection type is causal connection; And binding the disambiguated technical field information and the industry direction information to corresponding requirement sub-chains respectively to form an enterprise requirement feature set.
4. The method for matching talents of scientific research with information of enterprises according to claim 3, wherein constructing a result feature set comprises: The scientific research result information is disassembled into a result fragment set, and the result fragment set at least comprises a technical brief introduction fragment, a verification condition fragment, a verification result fragment and an application direction fragment; constructing a result evidence chain based on the result fragment set, wherein the result evidence chain is connected with the application pointing fragments according to the sequence of the technical profile fragments, the verification condition fragments and the verification result fragments; generating result maturity evaluation information based on a result evidence chain, wherein the result maturity evaluation information is determined by a verification condition segment and a verification result segment together, marking the result maturity evaluation information as an incomplete maturity state under the condition that the verification condition segment has a test environment constraint and the verification result segment lacks corresponding index description, and writing the incomplete maturity state into a result feature set; And generating an application scene prediction result based on the application pointing segment, wherein the application scene prediction result at least comprises three scene elements of an application object, a deployment form and an operation constraint.
5. The method for matching talents to enterprise information of claim 4, wherein constructing the set of enterprise technical layout features and identifying potential technical demand directions comprises: extracting a technical keyword set from historical technical information, and forming an enterprise technical evolution sequence according to time sequence; Dividing an enterprise technical evolution sequence into continuous evolution fragment groups, wherein each evolution fragment group corresponds to a technical direction aggregation result; comparing each demand sub-chain in the enterprise demand characteristic set with the evolution fragment group to obtain a demand coverage item set and a demand gap item set, wherein the demand gap item set is a technical direction or scene element which exists in the demand sub-chain but does not exist in the evolution fragment group; The method comprises the steps of decomposing a demand gap item set into a first type gap and a second type gap according to gap sources, wherein the first type gap is a technical direction gap, the second type gap is a scene element gap, and the first type gap and the second type gap are determined to be potential technical demand directions.
6. The method for matching talents to enterprise information of claim 5, wherein constructing the talent-to-scientific-research capability feature set comprises: acquiring scientific research achievements or patent information in the talent information of scientific research, constructing a corresponding achievements evidence chain or patent evidence chain aiming at each scientific research achievements or patent information, and extracting technical direction elements, verification elements and scene elements in the chain; merging a plurality of evidence chains corresponding to the scientific research talents according to scene elements to form a scientific research talent scene cluster set; Generating a capability indication vector for each scientific research talent scene cluster set, wherein the capability indication vector at least comprises a direction stability component and a verification completeness component, the direction stability component is determined by the consistency of technical direction elements in the scene cluster, and the verification completeness component is determined by the uniformity degree of verification elements; and fusing the capability indication vector with scientific research direction information to form a scientific research person capability feature set.
7. The method of claim 6, wherein mapping the set of enterprise demand features, the set of achievement features, the set of enterprise technical layout features, and the set of talent ability features to a unified feature space comprises: Mapping each demand sub-chain in the enterprise demand characteristic set into a demand chain vector, and reserving chain segment boundary indexes of the demand chain vector; Mapping the result evidence chain in the result feature set into a result chain vector, and reserving a verification segment boundary index of the result chain vector; Mapping capability indication vectors in the scientific research talent capability feature set into talent chain vectors, and reserving scene cluster boundary indexes of the talent chain vectors; In the unified feature space, a chain segment boundary index and a verification segment boundary index are aligned to be used as constraints to form a comparable chain alignment structure for subsequent matching analysis.
8. The method of matching talents to enterprise information of claim 7, wherein the matching analysis and generating a set of matching results comprises: segment matching is carried out on the scientific research person capacity vector and the enterprise demand vector based on the chain alignment structure, and the segment matching comprises direction segment matching, scene segment matching and verification segment matching; When the direction segment matching is successful and the scene segment matching is successful and the segment matching verification fails, the corresponding scientific research talents are brought into a supplementary scientific research talent set, and a supplementary reason mark is generated, wherein the supplementary reason mark at least comprises a maturity insufficient mark and an evidence missing mark; when the direction segment matching is successful and the scene segment matching is failed, a second type gap in the potential technical requirement direction is called to carry out playback verification on the scene segment, the playback verification is that the scene segment matching is re-executed by taking the second type gap as a substitute scene element, and when the playback verification is successful, the corresponding scientific research talents are brought into a potential cooperative scientific research talent set; when the direction segment matching fails, the corresponding scientific research talents are directly removed, and the matching result set is not entered.
9. The method of claim 8, wherein outputting the talent information and generating the associated recommendation information comprises: Generating matching basis description information aiming at each scientific research talent in the matching result set, wherein the matching basis description information at least comprises a hit requirement sub-chain index, a hit achievement evidence chain index, a hit talent scene cluster index and a corresponding supplementary reason mark or playback verification mark; outputting the matching basis description information and the scientific research talent information together; Receiving feedback information of enterprises on output results, and respectively updating a priority combination library and an exclusion combination library, wherein the priority combination library is used for recording effective pairing relations between a demand subchain index and a talent scene cluster index, and the exclusion combination library is used for recording ineffective pairing relations between the demand subchain index and the talent scene cluster index; In the subsequent matching analysis, the effective pairing relation in the priority combination library is preferentially called to generate a candidate scientific research person set, and the ineffective pairing relation recorded in the combination library is removed and excluded in a candidate screening stage.
10. A system for matching talents of scientific research with enterprise information, for implementing a method for matching talents of scientific research with enterprise information according to any one of claims 1 to 9, comprising: The first acquisition processing module is configured to acquire enterprise demand information of a target enterprise, wherein the enterprise demand information at least comprises technical field information, industry direction information, application scene description information and technical maturity requirement information, and performs semantic analysis and structuring processing on the enterprise demand information to construct an enterprise demand feature set; The second acquisition processing module is configured to acquire the information of the achievements of the scientific research, wherein the information of the achievements of the scientific research at least comprises technical profile information of the achievements, technical field information of the achievements, technical maturity evaluation information and technical development trend information, analyze the information of the achievements of the scientific research, construct an achievements feature set and generate an application scene prediction result based on the achievements feature set; The historical information processing module is configured to acquire historical technical information of a target enterprise, wherein the historical technical information at least comprises laid technical keyword information, existing technical direction information and published scientific research achievements or patent information of the enterprise, analyze the historical technical information, construct an enterprise technical layout feature set and identify potential technical demand directions based on the enterprise technical layout feature set; The third acquisition processing module is configured to acquire scientific research talent information, wherein the scientific research talent information at least comprises scientific research direction information, scientific research achievements or patent information and scientific research experience information, analyzes the scientific research talent information and constructs a scientific research talent capability feature set; The matching recommendation module is configured to map the enterprise demand feature set, the achievement feature set, the enterprise technical layout feature set and the scientific research person capacity feature set to a unified feature space to form an enterprise demand vector, an enterprise technical layout vector and a scientific research person capacity vector; Based on the unified feature space, carrying out matching analysis on the talent capability vector of the scientific research person and the enterprise demand vector and the enterprise technical layout vector to generate a matching result set between the scientific research person and a target enterprise; And outputting the talent information of the scientific research matched with the target enterprise based on the matching result set, and generating the recommendation information of the scientific research achievements or the cooperation docking information associated with the talent information of the scientific research.

Description

Scientific research talent and enterprise information matching method and system Technical Field The invention relates to the technical field of information technology, in particular to a method and a system for matching talents of scientific researches with enterprise information. Background Along with the acceleration and evolution of global technological innovation, the dependence of industrial technology upgrading on the conversion of scientific research achievements is continuously deepened, enterprises face urgent demands of technology iteration and product innovation, and a large number of scientific research talents and achievements thereof are dispersed in universities and research institutions to form a remarkable information barrier. The current scientific research achievement docking platform can intensively display enterprise requirements, scientific research achievements and talent information, but has fundamental defects in a matching mechanism. The existing system generally adopts a keyword matching or label mapping mode, cuts enterprise requirements, scientific research achievements and talent information into isolated data points, fails to construct a structural expression system of a logic chain, an achievement verification path chain and a talent capability evolution chain formed by requirements, and leads to the fact that matching results only reflect shallow semantic association and actual cooperative feasibility cannot be estimated. For example, technical requirements imposed by enterprises often imply multi-layer logical relationships, such as causal constraints of technical objects and application scenarios, relevance of maturity requirements and verification conditions, whereas traditional methods cannot resolve such structured semantics, resulting in recommendation results that deviate from real technical landing scenarios. Meanwhile, the enterprise technical layout has dynamic evolution characteristics, future technical notches are hidden in a historical technical keyword sequence, but the existing scheme lacks time sequence analysis capability on the enterprise technical evolution segment, and potential demand directions are difficult to reversely deduce from historical patents and scientific research achievements, so that high-value scientific research talents conforming to the strategic development directions of enterprises are systematically omitted. In the talent assessment link of scientific research, surface indexes such as scientific research direction quantity or patent counting are excessively relied on, comprehensive quantification of technical direction stability, verification element completeness and application scene consistency is ignored, so that talent capability image distortion can not distinguish short-term hot spot research from sustainable technical deep ploughing capability. In addition, the matching result output lacks structural basis, the enterprise is difficult to trace back key evidence chain nodes in the recommendation logic, and the user feedback information is not included in the closed-loop optimization mechanism, so that the matching accuracy cannot be improved through historical pairing relation accumulation, and the technological achievement conversion efficiency is restricted for a long time. Disclosure of Invention In view of the above, the invention provides a method and a system for matching talents in scientific research with enterprise information, which aims to effectively analyze the structural semantics of enterprise requirements, scientific research achievements and talent information, construct a unified feature space for accurate matching, and further improve the matching efficiency and the cooperation feasibility between the scientific research talents and the enterprise. In one aspect, the invention provides a method for matching talents of scientific research with enterprise information, which comprises the following steps: Acquiring enterprise demand information of a target enterprise, wherein the enterprise demand information at least comprises technical field information, industry direction information, application scene description information and technical maturity requirement information, carrying out semantic analysis and structuring processing on the enterprise demand information, and constructing an enterprise demand feature set; Acquiring scientific research result information, wherein the scientific research result information at least comprises result technology profile information, technical field information of a result, technical maturity evaluation information and technical development trend information, analyzing the scientific research result information, constructing a result feature set, and generating an application scene prediction result based on the result feature set; acquiring historical technical information of a target enterprise, wherein the historical technical information at least comprises laid tech