CN-122019882-A - Service demand matching method, device, storage medium and program product

CN122019882ACN 122019882 ACN122019882 ACN 122019882ACN-122019882-A

Abstract

The embodiment of the application provides a service demand matching method, equipment, a storage medium and a program product. The method includes the steps of collecting user service requirement information through dialogue interaction nodes in a natural language multi-round interaction mode, extracting structured requirement data from the requirement information through code execution nodes, calling external matching engine service through HTTP request nodes, executing hard constraint, skill and space-time multi-dimensional matching calculation on the structured requirement data based on a pre-constructed service provider knowledge graph to obtain a matching result, generating recommendation information through a template generation node, and returning the recommendation information to a user through an answer node. The flow of natural language interaction, demand analysis and multidimensional matching is realized based on the dialogue workflow, so that the accuracy and efficiency of matching are improved while the natural convenience of user interaction is ensured.

Inventors

LI YUJING
ZHANG JIAYI
SUN BO
LIU YAJING
GAO HUINING
LIU JING
LI WEINA
LI SIQING

Assignees

北京五八信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (12)

1. The service demand matching method is suitable for a server device, wherein the server device is preconfigured with a dialogue workflow system, the dialogue workflow system comprises a dialogue interaction node, a code execution node, an HTTP request node, a templated generation node and an answer node, and the method comprises the following steps: through the dialogue interaction node, carrying out multi-round interaction with a user in a natural language dialogue mode so as to collect service demand information of the user; Extracting, by the code execution node, structured demand data for at least one demand dimension from the service demand information; Invoking an external matching engine service through the HTTP request node; In the external matching engine service, based on a pre-constructed knowledge graph of a service provider, performing multi-dimensional matching calculation on the structured demand data to obtain a matching result, wherein the multi-dimensional matching at least comprises hard constraint matching, skill matching and space-time matching; and generating recommendation information according to the matching result through the templated generating node, and returning the recommendation information to the user through the answer node.
2. The method of claim 1, wherein the dialogue interaction node comprises a large language model node and a condition judgment node; through the dialogue interaction node, carrying out multi-round interaction with a user in a natural language dialogue mode so as to collect service requirement information of the user, wherein the method comprises the following steps: Generating and outputting a target question through the large language model node in response to the initial demand input by the user; Determining the completeness of the answer of the user to the target question through the condition judging node; And if the information integrity of the answer is not higher than a preset integrity threshold, generating and outputting a new target question through the large language model node until the condition judgment node judges that the integrity of the answer of the user to the new target question is higher than the preset integrity threshold, and generating service demand information according to the answer of the user to the new target question.
3. The method of claim 1, wherein extracting structured demand data for at least one demand dimension from the service demand information comprises: identifying candidate information of at least one requirement dimension from the service requirement information, wherein the candidate information comprises equipment type, abnormal phenomenon, position information, time requirement and/or certificate information; Carrying out standardization processing on the candidate information to obtain standardized data, and identifying safety constraint conditions existing in the standardized data; And packaging the standardized data and the security constraint condition into the structured demand data in a target format.
4. A method according to claim 3, wherein normalizing the candidate information to obtain normalized data comprises: Mapping the candidate information to a predefined standardized class to obtain candidate standardized data; In response to identifying that the location information contains floor information and the floor is higher than a preset threshold, adding a safety constraint condition corresponding to the aerial work in the candidate standardized data; in response to identifying that the time requirement contains a keyword of emergency degree, adding a classification label corresponding to the emergency requirement in the candidate standardized data; In response to identifying that a particular certificate name is included in the certificate information, attribute information associated with the certificate name, common tools and/or service characteristics, is added to the candidate standardized data to obtain the standardized data.
5. The method of claim 1, wherein performing a multi-dimensional matching calculation on the structured demand data based on a pre-built knowledge-graph of a service provider to obtain a matching result comprises: Taking the service provider meeting the security constraint condition in the structured demand data as a first service provider matched through hard constraint from the service provider knowledge graph, wherein the number of the first service provider is one or more; performing skill matching on the first service provider and the structured demand data to obtain a first matching degree of the skill of the first service provider and the abnormal phenomenon in the structured demand data; performing space-time matching on the first service provider and the structured demand data to obtain a second matching degree of the spatial attribute information of the first service provider and the structured demand data; calculating a target matching score according to the first matching degree and the second matching degree; And screening target service providers which are subjected to multidimensional matching with the structured demand data from the first service provider according to the target matching score so as to obtain a matching result.
6. The method of claim 5, wherein from the service provider knowledge graph, taking a service provider that satisfies security constraints in the structured demand data as a first service provider that is matched by hard constraints, comprising: Under the condition that the structured demand data contains safety constraint conditions corresponding to the aerial work, inquiring whether any service provider in the service provider knowledge graph is associated with effective aerial work license information; if not, the any service provider does not meet the safety constraint condition, and if so, the any service provider is the first service provider meeting the safety constraint condition.
7. The method of claim 5, wherein the service provider knowledge base comprises a plurality of skill nodes for each first service provider; Performing skill matching on the first service provider and the structured demand data to obtain a first matching degree of the skills of the first service provider and the abnormal phenomenon in the structured demand data, wherein the skill matching comprises the following steps: Generating a plurality of skill requirements according to abnormal phenomena in the structured demand data; Determining a plurality of target skill nodes matched with semantic information of the skill requirements from a plurality of skill nodes of each first service provider; And carrying out weighted calculation on the candidate matching degree of the attribute information between each target skill node and the matched skill requirement thereof to obtain a first matching degree between the skill of each first service provider and the abnormal phenomenon in the structured demand data.
8. The method of claim 5, wherein the service provider knowledge graph comprises service scope nodes for each first service provider; Performing space-time matching on the first service provider and the structured demand data to obtain a second matching degree of spatial attribute information of the first service provider and the structured demand data, including: Calculating the geographic distance and service range conformity of the user and each first service provider based on the spatial attribute information of the structured demand data and the spatial attribute information of the service range node of each first service provider; Calculating the response time of each first service provider based on the current service state and queuing situation of each first service provider; a second degree of matching of each first service provider to spatial attribute information of the structured demand data is calculated based on the geographic distance, the service range compliance, and the response time.
9. The method of claim 5, wherein generating recommendation information based on the matching result comprises: selecting a target interpretation template from predefined interpretation templates based on the matching result; Filling the detail information of each target service provider into a target interpretation template according to the priority order of each target service provider to obtain candidate information; Generating action suggestion information according to the detail information of each target service provider, and adding the action suggestion information into the candidate information to obtain recommendation information, wherein the action suggestion information comprises a selection mode, a contact mode and/or an adjustment requirement option.
10. A server device comprising a memory for storing one or more computer instructions and a processor for executing the one or more computer instructions for performing the steps of the method of any of claims 1-9.
11. A computer readable storage medium, characterized in that a computer program, when executed by a processor, causes the processor to carry out the steps of the method of any one of claims 1-9.
12. A computer program product comprising computer programs/instructions which, when executed by a processor, cause the processor to carry out the steps of the method of any of claims 1-9.

Description

Service demand matching method, device, storage medium and program product Technical Field The present application relates to the field of computer technologies, and in particular, to a service requirement matching method, device, storage medium, and program product. Background With the wide application of online service platforms, service demand matching is used as a core link for connecting users and service resources, and the efficiency and quality of the service demand matching directly affect the user experience and the platform operation efficiency. The current mainstream matching technology mainly depends on two types of implementation modes, namely firstly, a user submits requirements through a preset structured form (such as fields of service type, time, place and the like), the system performs matching based on a rule base, and secondly, the user inputs free text keywords, and the system returns to a service list through keyword retrieval. However, the structured form interaction process is mechanically stiff, requires users to define various parameters in advance, is difficult to adapt to ambiguity and diversity of natural language expression, is easy to cause missing of required information or increase of operation burden of the users, and the keyword retrieval mode is low in interaction threshold, but can only realize shallow literal matching, lacks deep analysis capability on semantic connotation and implicit conditions of the user requirements, and is obviously insufficient in reliability and applicability of matching results when the actual scene of multiple factors is comprehensively considered. In addition, in the prior art, the requirement acquisition and matching calculation links often have the splitting treatment, and a dynamic collaboration mechanism is lacked, so that the inherent contradiction between interaction experience and matching quality is further enhanced. In summary, the existing service demand matching method is difficult to effectively improve the understanding depth of the demand intention and the accuracy of the matching result in the matching process while ensuring natural and convenient user interaction. Disclosure of Invention Aspects of the present application provide a service requirement matching method, apparatus, storage medium, and program product, for improving accuracy and efficiency of matching while ensuring natural and convenient user interaction. The embodiment of the application provides a service demand matching method which is suitable for service side equipment, wherein the service side equipment is pre-configured with a dialogue workflow system, the dialogue workflow system comprises dialogue interaction nodes, code execution nodes, HTTP request nodes, template generation nodes and answer nodes, the method comprises the steps of conducting multi-round interaction with a user in a natural language dialogue mode through the dialogue interaction nodes to collect service demand information of the user, extracting structured demand data of at least one demand dimension from the service demand information through the code execution nodes, calling external matching engine service through the HTTP request nodes, executing multidimensional matching calculation on the structured demand data based on a pre-constructed service provider knowledge graph in the external matching engine service to obtain a matching result, generating recommendation information according to the matching result through the template generation nodes, and returning the recommendation information to the user through the answer nodes. The dialogue interaction node comprises a large language model node and a condition judgment node, wherein the dialogue interaction node is used for carrying out multi-round interaction with a user in a natural language dialogue mode to collect service requirement information of the user, the dialogue interaction node comprises the steps of responding to initial requirements input by the user, generating and outputting target questions through the large language model node, determining the completeness of answers of the user to the target questions through the condition judgment node, generating and outputting new target questions through the large language model node if the information completeness of the answers is not higher than a preset completeness threshold, and generating service requirement information according to the answers of the user to the new target questions until the condition judgment node is used for judging that the completeness of the answers of the user to the new target questions is higher than the preset completeness threshold. Optionally, the method comprises the steps of extracting structured demand data of at least one demand dimension from the service demand information, wherein the structured demand data comprises candidate information of at least one demand dimension from the service demand information, the candidate informati