CN-116932841-B - Resource query method, device, server and storage medium

CN116932841BCN 116932841 BCN116932841 BCN 116932841BCN-116932841-B

Abstract

The present disclosure relates to a resource query method, apparatus, server, storage medium, and computer program product. The method comprises the steps of responding to a resource query request, obtaining reference characteristics and attribute conditions of resources to be queried, filtering resources to be filtered in associated resources of initial resources in a comprehensive attribute communication graph according to the attribute conditions, screening target resources matched with the reference characteristics from the filtered associated resources, taking the target resources as new initial resources, jumping to the resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph according to the attribute conditions, filtering the resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph, screening the target resources matched with the reference characteristics from the filtered associated resources until the new target resources cannot be screened, and obtaining a resource query result corresponding to the resource query request according to each screened target resource. By adopting the method, the resource query efficiency can be improved.

Inventors

WANG YU
Mao Qiuli
LIU JUN
LIU LI
DAI GUOHAO

Assignees

清华大学
北京达佳互联信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20230710

Claims (10)

1. A method for querying resources, comprising: the method comprises the steps of responding to a resource query request, acquiring reference characteristics of a resource to be queried and attribute conditions, wherein the reference characteristics refer to account characteristics of a request account of the resource query request, and the attribute conditions refer to resource attributes serving as query conditions, including new uploading and high heat; Filtering the resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph according to the attribute conditions, and screening target resources matched with the reference characteristics from the filtered associated resources, wherein the comprehensive attribute communication graph is obtained by fusion processing of a full attribute communication graph and a same attribute communication graph of each resource attribute, the full attribute communication graph is used for describing the association relationship between the resources of each resource attribute, the same attribute communication graph is used for describing the association relationship between the resources of the same resource attribute, the initial resources are any one of the resources meeting the attribute conditions in the comprehensive attribute communication graph, the resources to be filtered are the resources which do not meet the attribute conditions in the associated resources, and the filtered associated resources comprise the resources except the resources to be filtered in the associated resources of the initial resources and the left resources after the filtering of the resources to be filtered; Taking the target resource as a new initial resource, and jumping to the step of filtering the resources to be filtered in the associated resources of the initial resource in the comprehensive attribute communication graph according to the attribute condition, and screening the target resource matched with the reference feature from the filtered associated resources until the new target resource cannot be screened from the comprehensive attribute communication graph; obtaining a resource query result corresponding to the resource query request according to each screened target resource; Filtering the resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph according to the attribute conditions, and screening the target resources matched with the reference characteristics from the filtered associated resources, wherein the method comprises the following steps: Determining the filtering probability of resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph according to the number of the attribute conditions and the number of the resource attributes related to the comprehensive attribute communication graph, carrying out corresponding filtering processing on the resources to be filtered in the associated resources according to the filtering probability to obtain the remaining resources in the resources to be filtered, obtaining filtered associated resources according to the resources except the resources to be filtered in the associated resources and the remaining resources in the resources to be filtered, and screening out target resources matched with the reference characteristics from the filtered associated resources.
2. The method of claim 1, further comprising, prior to obtaining the reference characteristics and attribute conditions for the resource to be queried in response to the resource query request: Acquiring a preset resource set, wherein the preset resource set comprises resource sets of all resource attributes; Respectively constructing a same-attribute communication graph of each resource attribute according to the similarity between the resources in the resource set of each resource attribute, and constructing the full-attribute communication graph according to the similarity between the resources in the preset resource set; And carrying out fusion processing on the full-attribute communication graph and the same-attribute communication graph of each resource attribute to obtain the comprehensive attribute communication graph.
3. The method of claim 1, further comprising, prior to filtering the resources to be filtered in the associated resources of the initial resources in the integrated attribute connectivity graph according to the attribute conditions: Determining initial resources in the comprehensive attribute communication graph from resources meeting the attribute conditions in the comprehensive attribute communication graph; Determining the associated resource of the initial resource from the comprehensive attribute communication graph; and screening out the resources which do not meet the attribute conditions from the associated resources, and taking the resources as the resources to be filtered.
4. The method of claim 1, wherein the screening the filtered associated resources for the target resources that match the reference feature comprises: determining, for each resource in the filtered associated resources, a similarity between a resource feature of the resource and the reference feature; and screening out the resources with the similarity meeting the first similarity condition from the filtered associated resources, and taking the resources as target resources matched with the reference characteristics.
5. The method according to any one of claims 1 to 4, wherein obtaining a resource query result corresponding to the resource query request according to each of the screened target resources includes: Screening out the resources with the similarity meeting the second similarity condition from the screened target resources; And determining a resource query result corresponding to the resource query request according to the screened resources with the similarity meeting the second similarity condition between the resource query request and the reference feature.
6. The method of claim 5, wherein the obtaining, in response to the resource query request, the reference feature and the attribute condition of the resource to be queried comprises: analyzing the resource query request to obtain a request account of the resource query request and attribute conditions selected by the request account; the account characteristics of the request account are obtained to be used as reference characteristics of the resource to be queried, and the attribute conditions selected by the request account are used as the attribute conditions of the resource to be queried.
7. A resource query device, comprising: The resource query system comprises a request processing unit, a resource query unit and a resource query unit, wherein the request processing unit is configured to execute a request for responding to a resource query request and acquire reference characteristics and attribute conditions of a resource to be queried, the reference characteristics refer to account characteristics of a request account of the resource query request, and the attribute conditions refer to resource attributes serving as query conditions, wherein the resource attributes comprise new uploading and high heat; The resource filtering unit is configured to perform filtering on resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph according to the attribute conditions, and select target resources matched with the reference characteristics from the filtered associated resources, wherein the comprehensive attribute communication graph is obtained by fusion processing of a full attribute communication graph and the same attribute communication graph of each resource attribute, the full attribute communication graph is used for describing the association relationship among the resources of each resource attribute, the same attribute communication graph is used for describing the association relationship among the resources of the same resource attribute, the initial resources are any one of the resources meeting the attribute conditions in the comprehensive attribute communication graph, the resources to be filtered are the resources which do not meet the attribute conditions in the associated resources, the filtered associated resources comprise the resources except the resources to be filtered in the associated resources of the initial resources, and the remaining resources after the filtering of the resources to be filtered; The resource screening unit is configured to perform the steps of taking the target resource as a new initial resource, jumping to the filtering resource to be filtered in the associated resources of the initial resource in the comprehensive attribute communication graph according to the attribute condition, and screening the target resource matched with the reference feature from the filtered associated resources until the new target resource cannot be screened from the comprehensive attribute communication graph; the result determining unit is configured to execute the resource query results corresponding to the resource query requests according to the screened target resources; The resource filtering unit is further configured to determine the filtering probability of resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph according to the number of the attribute conditions and the number of the resource attributes related to the comprehensive attribute communication graph, perform corresponding filtering processing on the resources to be filtered in the associated resources according to the filtering probability to obtain the remaining resources in the resources to be filtered, obtain the filtered associated resources according to the resources except the resources to be filtered in the associated resources and the remaining resources in the resources to be filtered, and screen out target resources matched with the reference characteristics from the filtered associated resources.
8. The apparatus according to claim 7, further comprising a graph construction unit configured to perform obtaining a preset resource set, wherein the preset resource set includes a resource set of each resource attribute, construct a same-attribute connected graph of each resource attribute according to a similarity between resources in the resource set of each resource attribute, and construct the full-attribute connected graph according to the similarity between resources in the preset resource set, and perform fusion processing on the full-attribute connected graph and the same-attribute connected graph of each resource attribute to obtain the comprehensive attribute connected graph.
9. A server for a server, which comprises a server and a server, characterized by comprising the following steps: A processor; a memory for storing the processor-executable instructions; Wherein the processor is configured to execute the instructions to implement the resource query method of any of claims 1 to 6.
10. A computer readable storage medium, characterized in that instructions in the computer readable storage medium, when executed by a processor of a server, enable the server to perform the resource query method of any one of claims 1 to 6.

Description

Resource query method, device, server and storage medium Technical Field The present disclosure relates to the field of computer technology, and in particular, to a resource query method, apparatus, server, storage medium, and computer program product. Background With the development of computer technology, resource searching technology has emerged, for example, by screening out resources satisfying a given condition from a large number of resources by a search algorithm. In the related art, when searching resources, the final resource query result is generally determined by calculating the similarity between the characterization vector of each resource of the resource library and the input target vector, but the number of resources in the resource library is very large, so that the calculation amount is large, and the resource query efficiency is low. Disclosure of Invention The present disclosure provides a resource query method, apparatus, server, storage medium, and computer program product, to at least solve the problem of low resource query efficiency in the related art. The technical scheme of the present disclosure is as follows: according to a first aspect of an embodiment of the present disclosure, there is provided a resource query method, including: responding to a resource query request, and acquiring reference characteristics and attribute conditions of a resource to be queried; filtering the resources to be filtered in the associated resources of the initial resources in the comprehensive attribute communication graph according to the attribute conditions, and screening target resources matched with the reference characteristics from the filtered associated resources, wherein the comprehensive attribute communication graph is obtained by fusion processing of a full attribute communication graph and an identical attribute communication graph of each resource attribute, the full attribute communication graph is used for describing the association relationship between the resources of each resource attribute, the identical attribute communication graph is used for describing the association relationship between the resources of the identical resource attribute, the initial resources are any one of the resources meeting the attribute conditions in the comprehensive attribute communication graph, and the resources to be filtered are the resources which do not meet the attribute conditions in the associated resources; Taking the target resource as a new initial resource, and jumping to the step of filtering the resources to be filtered in the associated resources of the initial resource in the comprehensive attribute communication graph according to the attribute condition, and screening the target resource matched with the reference feature from the filtered associated resources until the new target resource cannot be screened from the comprehensive attribute communication graph; and obtaining a resource query result corresponding to the resource query request according to each screened target resource. In an exemplary embodiment, before the obtaining of the reference feature and the attribute condition of the resource to be queried in response to the resource query request, the method further includes: Acquiring a preset resource set, wherein the preset resource set comprises resource sets of all resource attributes; Respectively constructing a same-attribute communication graph of each resource attribute according to the similarity between the resources in the resource set of each resource attribute, and constructing the full-attribute communication graph according to the similarity between the resources in the preset resource set; And carrying out fusion processing on the full-attribute communication graph and the same-attribute communication graph of each resource attribute to obtain the comprehensive attribute communication graph. In an exemplary embodiment, the filtering the resources to be filtered in the associated resources of the initial resources in the comprehensive attribute connectivity graph according to the attribute condition includes: Determining filtering parameters of resources to be filtered in associated resources of initial resources in the comprehensive attribute communication graph according to the number of the attribute conditions and the number of the resource attributes related to the comprehensive attribute communication graph; and filtering the resources to be filtered in the associated resources according to the filtering parameters to obtain the filtered associated resources. In an exemplary embodiment, before filtering the resources to be filtered in the associated resources of the initial resources in the integrated attribute connectivity graph according to the attribute condition, the method further includes: Determining initial resources in the comprehensive attribute communication graph from resources meeting the attribute conditions in the comprehensive attri