CN-121997072-A - Intelligent matching method and system for park-resident enterprises

CN121997072ACN 121997072 ACN121997072 ACN 121997072ACN-121997072-A

Abstract

The invention discloses a park-resident enterprise intelligent matching method and system, which relate to the technical field of data intelligent matching, and comprise the following steps of constructing an industry knowledge graph based on industry data, calculating a first matching degree of a target enterprise and a candidate park according to the industry knowledge graph and attribute information of the target enterprise, generating a structured demand vector of the target enterprise based on demand information of the target enterprise, generating a structured supply vector based on service supply information of the candidate park, calculating similarity between the demand vector and the supply vector to obtain a second matching degree of the target enterprise and the candidate park, comprehensively processing the service supply information of the candidate park, the first matching degree and the second matching degree to generate a comprehensive matching result, ensuring that the matching result maximizes double benefits of industry cooperation and service adaptation on the premise of meeting the rigidity demand of the enterprise, and improving the precision of a vendor, the satisfaction degree of the enterprise and the ecological quality of the park industry.

Inventors

XIE WEICHAO
HU RUIHANG
HUANG ZEFENG

Assignees

深圳市伙伴网络服务科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (10)

1. An intelligent matching method for a park to enter a resident enterprise is characterized by comprising the following steps, Step S1, an industry knowledge graph is constructed based on industry data, and a first matching degree between a target enterprise and a candidate park is calculated according to the industry knowledge graph and attribute information of the target enterprise; Step S2, a structured demand vector of a target enterprise is generated based on demand information of the target enterprise, and a structured supply vector is generated based on service supply information of a candidate park; calculating the similarity between the demand vector and the supply vector to obtain a second matching degree between the target enterprise and the candidate park; And S3, comprehensively processing the service supply information of the candidate parks and the first matching degree and the second matching degree to generate a comprehensive matching result.
2. The intelligent matching method for parks to stay in enterprises according to claim 1, wherein attribute information, demand information, industry data and service supply information of candidate parks of a target enterprise are obtained; the attribute information comprises the industry attribute and the business attribute of the target enterprise; the business attribute comprises the registered capital and the established years of the target enterprise; The demand information comprises a demand description of the service by the target enterprise; wherein the demand description includes demands for physical space conditions, policy services, and financial services; The requirements of the physical space conditions comprise a target enterprise requirement area, a target enterprise requirement net height and a target enterprise requirement load, and the target enterprise marks the necessary physical space conditions, policy services and financial services in the requirement description; The service offer information includes resource conditions for the candidate campus, service entries, industry scope supported, minimum capital requirement for residence; wherein the resource conditions include available physical space conditions for the candidate campus, the available physical space conditions including candidate campus rentable area of the candidate campus, candidate campus floor clearance, and candidate campus floor load; The service items include policy service, financial service, talent service, and marketing service, and the service items cover the requirements of spatial conditions, policy service, and financial service.
3. The intelligent matching method for parks to enter enterprises according to claim 2, wherein step S1, an industry knowledge graph is constructed based on industry data, and a first matching degree of a target enterprise and a candidate park in an industry ecological dimension is calculated according to the industry knowledge graph and attribute information of the target enterprise; the industrial data comprise enterprise industrial and commercial information data, industry classification standard data and macro economic relation data; Step S1 includes step S101, step S102, step S103, and step S104; step S101, based on industry classification standard data, constructing an industry classification node comprising a hierarchical structure; Based on macro economic relationship data, establishing an association relationship between industrial classification nodes, wherein the association relationship comprises an upstream and downstream supply chain relationship, a technical cooperation relationship or a service matching relationship; based on enterprise business information data, extracting enterprise entities and industry attributes, and creating corresponding enterprise entity nodes; Step S102, connecting the enterprise entity node to the corresponding industry classification node according to the subordinate relation between the enterprise entity node and the industry classification node; Integrating the association relationship among the industry classification nodes, the enterprise entity nodes and the industry classification nodes to form an industry knowledge graph; Step S103, mapping the industry attribute of the target enterprise into the industry knowledge graph to obtain a corresponding industry classification node serving as a target industry node; Mapping the industry attributes of all resident enterprises in the candidate parks into the industry knowledge graph to obtain a set formed by industry classification nodes corresponding to all resident enterprises, wherein the set is used as the industry classification node set of the candidate parks; step S104, calculating the shortest path length of the industrial knowledge graph between the target industrial node and each node in the industrial node set in the industrial knowledge graph; Based on the shortest path length, obtaining a corresponding industry association degree through a preset inverse proportion function; And based on the industrial association degree of the target industrial node and all nodes in the industrial classification node set, combining the weight of each node, and obtaining the first matching degree of the target enterprise and the candidate park through weighted calculation.
4. The intelligent matching method for a campus-resident enterprise of claim 3, wherein step S2, a structured demand vector of the target enterprise is generated based on demand information of the target enterprise, and a structured supply vector is generated based on candidate campus service supply information; step S2 includes step S201, step S202, step S203, step S204, and step S205; Step S201, constructing a feature list comprising a plurality of dimension matching features, wherein the matching features comprise physical space features, policy service features and financial service features; Setting standardized names, data types, a preset quantization value set and a regular expression extracted from characterization information for each matching feature; Step S202, matching the requirement description of the target enterprise by using a regular expression corresponding to each matching feature in the feature list; If the matching is successful and the text fragment is extracted, obtaining a value from a quantized set corresponding to the matching feature according to the extracted text fragment, and taking the value as a required intensity value on the matching feature; Step S203, a weight coefficient is allocated to each matching feature based on the business attribute of the target enterprise, and all the demand intensity values are weighted by the weight coefficient, so as to generate a structured demand vector of the target enterprise.
5. The intelligent matching method for parks to enter enterprises according to claim 4, wherein step S204, for each matching feature in the feature list, matches the service supply information of the candidate parks using the regular expression corresponding to the matching feature; If the matching is successful and the text fragment is extracted, obtaining a value from a quantization set corresponding to the matching feature according to the extracted text fragment, and taking the value as a supply capacity value on the matching feature; Step S205, normalizing the supply capability values of all candidate parks based on the maximum supply capability value of the same matching feature of all candidate parks, and generating a structured supply vector of each candidate park.
6. The intelligent matching method for a campus-resident enterprise of claim 5, wherein the second matching degree is obtained by calculating the demand vector and the supply vector through a cosine similarity formula.
7. The intelligent matching method for parks to enter enterprises according to claim 6, wherein step S3, the service supply information of the candidate parks, the first matching degree and the second matching degree are comprehensively processed to generate comprehensive matching results; step S3 includes step S301, step S302, and step S303; Step S301, judging a recommendation level according to the first matching degree and the second matching degree based on the history matching data and the campus recruitment strategy; recommendation levels include priority recommendation, considered and not recommended; the recommendation level judging logic is used for judging that the recommendation is preferentially recommended if the first matching degree is larger than or equal to a first threshold value and the second matching degree is larger than or equal to a second threshold value; If the first matching degree is greater than or equal to a third threshold value or the second matching degree is greater than or equal to a fourth threshold value, judging that the matching degree is considered; if the first matching degree is equal to or less than the fifth threshold value and the second matching degree is equal to or less than the sixth threshold value, the recommendation is judged not to be made.
8. The intelligent matching method for parks to enter enterprises according to claim 7, wherein step S302, candidate parks are screened based on preset constraint conditions, and candidate parks which do not meet any constraint condition are removed to obtain a final candidate park set; constraint conditions include physical space constraints, qualification adaptation constraints and service coverage constraints; The physical space constraint comprises that the target enterprise demand area is smaller than or equal to the rentable area of the candidate park, the floor clearance of the candidate park is larger than or equal to the target enterprise demand clearance, and the floor load of the candidate park is larger than or equal to the target enterprise demand load; The qualification adaptation constraint is that the target enterprise camping industry classification code belongs to the industry range supported by the candidate park and the registered capital of the target enterprise is greater than or equal to the minimum resident capital requirement of the candidate park; The service coverage constraint is that service entries of the candidate campus are marked as necessary physical space conditions for the target enterprise and coverage of the policy service and the financial service is greater than or equal to a coverage threshold.
9. The intelligent matching method for parks to enter enterprises according to claim 8, wherein in step S303, if the final candidate parks set is empty, a comprehensive matching result of the non-recommended parks is generated; If the final candidate park set is not empty, dividing the candidate parks in the final candidate park set according to the recommendation level determined in the step S301; In the same recommended level, sorting in descending order according to the weighted summation results of the first matching degree and the second matching degree; if the weighted summation results are the same, sorting in descending order according to the first matching degree; And if the weighted summation result and the first matching degree are the same, the coverage rate of the demand which is not marked as necessary for the target enterprise is sorted in a descending order according to the service items of the candidate parks, and the sorted candidate parks list is output as the comprehensive matching result.
10. A park resident enterprise intelligent matching system applied to the park resident enterprise intelligent matching method as claimed in any one of claims 1-9, and characterized by comprising a construction module, a calculation module and a matching module; The construction module is used for constructing an industry knowledge graph based on the industry data and calculating a first matching degree between the target enterprise and the candidate park according to the industry knowledge graph and attribute information of the target enterprise; the computing module is used for generating a structured demand vector of the target enterprise based on the demand information of the target enterprise and generating a structured supply vector based on the service supply information of the candidate park; calculating the similarity between the demand vector and the supply vector to obtain a second matching degree between the target enterprise and the candidate park; And the matching module is used for comprehensively processing the service supply information, the first matching degree and the second matching degree of the candidate parks to generate a comprehensive matching result.

Description

Intelligent matching method and system for park-resident enterprises Technical Field The invention relates to the technical field of intelligent data matching, in particular to an intelligent matching method and system for a park to enter a resident enterprise. Background Along with the large-scale and specialized development of industrial parks in China in recent years, the accurate matching of the parks and the enterprise site selection becomes a key link for improving the operation efficiency of the parks and promoting the industrial aggregation, the data intelligent matching technology has great potential in solving the problem of the supply and demand butt joint, and by introducing a method of big data analysis, a knowledge graph and machine learning, the method aims at replacing the traditional matching mode relying on manual experience and rough recommendation, realizing rapid screening and accurate recommendation from massive information, and the importance of the technology is embodied in the aspects of improving the success rate of a recruitment and the satisfaction of enterprises, and promoting the optimization of the ecological environment of the park industry and the cooperation of an industry chain through scientific analysis, so that the economic high-quality development of a region is driven. However, the existing matching technology still has significant limitations, on one hand, most methods simply screen based on enterprise classification and scale shallow attributes, lack deep modeling of ecological associations such as complex upstream and downstream, technical cooperation and the like among industries, and cause difficulty in evaluating industrial cooperation potential after the enterprises reside, on the other hand, the existing technology generally compares enterprise demands with campus supplies as isolated labels, and fails to convert unstructured demand description and supply information into quantitatively calculated structured vectors, so that the matching degree between service supplies and demands cannot be measured finely, and core appeal of the enterprises and characteristic service capability of the campus are easily ignored. Disclosure of Invention The technical problems solved by the invention are that the prior matching technology still has significant limitation, on one hand, most methods are simply screened based on enterprise classification and scale shallow attributes, and lack of deep modeling on complex upstream and downstream, technical cooperation and other ecological associations among industries, so that the industrial cooperative potential after the enterprise is resided is difficult to evaluate, on the other hand, the prior art generally compares enterprise demands with campus supplies as isolated labels, and fails to convert unstructured demand description and supply information into quantitatively calculated structured vectors, so that the matching degree between service supplies and demands cannot be finely measured, and the core appeal of the enterprise and the characteristic service capability of the campus are easy to ignore. In order to solve the technical problems, the invention provides the following technical proposal that the intelligent matching method for the park-resident enterprises comprises the following steps, Step S1, an industry knowledge graph is constructed based on industry data, and a first matching degree between a target enterprise and a candidate park is calculated according to the industry knowledge graph and attribute information of the target enterprise; Step S2, a structured demand vector of a target enterprise is generated based on demand information of the target enterprise, and a structured supply vector is generated based on service supply information of a candidate park; calculating the similarity between the demand vector and the supply vector to obtain a second matching degree between the target enterprise and the candidate park; and S3, comprehensively processing the service supply information, the first matching degree and the second matching degree of the candidate parks to generate a comprehensive matching result. As a preferable scheme of the intelligent matching method for the park-resident enterprises, the invention obtains attribute information, demand information, industry data and service supply information of candidate parks of target enterprises; the attribute information comprises the industry attribute and the business attribute of the target enterprise; the business attribute comprises the registered capital and the established years of the target enterprise; The demand information comprises a demand description of the service by the target enterprise; wherein the demand description includes demands for physical space conditions, policy services, and financial services; The requirements of the physical space conditions comprise a target enterprise requirement area, a target enterprise requirement net height