CN-121981128-A - Enterprise knowledge question and answer method and device, computer equipment and storage medium

CN121981128ACN 121981128 ACN121981128 ACN 121981128ACN-121981128-A

Abstract

The application relates to an enterprise knowledge question and answer method, an enterprise knowledge question and answer device, computer equipment and a storage medium. The enterprise knowledge questioning method comprises the steps of responding to acquired enterprise knowledge questioning content, carrying out vectorization processing on the enterprise knowledge questioning content to obtain enterprise knowledge questioning vectors, determining target sub-segments matched with the enterprise knowledge questioning content from the sub-segment sets according to the enterprise knowledge questioning vectors and feature vectors of all the sub-segments in the sub-segment sets, determining target father segments associated with the target sub-segments from father segments associated with all the sub-segments in the sub-segment sets, and generating and outputting enterprise knowledge answer content aiming at the enterprise knowledge questioning content according to the enterprise knowledge questioning content, the target sub-segments and the target father segments. By adopting the method, the query efficiency of enterprise knowledge can be improved.

Inventors

Zhu Yanxiong
WU TAO
ZHAO YAN
LIN FENG
WANG NAN
ZHU YANSHENG
XU XIAOKAI

Assignees

深圳市爱都科技有限公司

Dates

Publication Date: 20260505
Application Date: 20251230

Claims (10)

1. An enterprise knowledge question and answer method, wherein the method is applied to an enterprise knowledge question and answer system, and the method comprises: Responding to the acquired enterprise knowledge questioning content, and carrying out vectorization processing on the enterprise knowledge questioning content to obtain an enterprise knowledge questioning vector; Determining target sub-segments matched with the enterprise knowledge question content from the sub-segment set according to the enterprise knowledge question vector and the feature vector of each sub-segment in the sub-segment set, wherein the sub-segment set comprises at least one sub-segment associated with each father segment in a father segment set, and the at least one sub-segment associated with each father segment is obtained by carrying out segment processing on each enterprise knowledge text in an enterprise knowledge text base; determining a target parent segment associated with the target sub-segment from parent segments associated with each sub-segment in a sub-segment set; And generating and outputting enterprise knowledge answer content aiming at the enterprise knowledge questioning content according to the enterprise knowledge questioning content, the target sub-segment and the target father segment.
2. The method of claim 1, wherein the determining the target sub-segment from the set of sub-segments that the enterprise knowledge question matches based on the enterprise knowledge question vector and the feature vector of each sub-segment in the set of sub-segments comprises: Determining feature vectors of a plurality of sub-segments to be selected according to the similarity between the feature vectors of the sub-segments in the sub-segment set and the enterprise knowledge question vector; Determining the matching degree between the feature vector of each sub-segment to be selected and the enterprise knowledge question vector according to the similarity between the feature vector of each sub-segment to be selected and the enterprise knowledge question vector and according to the validity weight of each sub-segment to be selected, wherein the validity weight of each sub-segment to be selected is included in the validity weight of each sub-segment in the sub-segment set, and the validity weight of each sub-segment in the sub-segment set is determined according to the data source of each sub-segment in the sub-segment set, the disclosure time of each sub-segment in the sub-segment set and the version of each sub-segment in the sub-segment set; And determining target subsections matched with the enterprise knowledge question content from the subsections set according to the matching degree between the feature vector of each subsection to be selected and the enterprise knowledge question vector.
3. The method according to claim 2, wherein the method further comprises: Acquiring accuracy feedback information of the enterprise knowledge answer content; Determining the current feedback accuracy of the target sub-segment according to the accuracy feedback information; And adjusting the effectiveness weight of the target sub-segment according to the current feedback accuracy of the target sub-segment, the working life of the current feedback person corresponding to the target sub-segment, the historical feedback accuracy of the target sub-segment and the working life of the historical feedback person of the target sub-segment.
4. The method of claim 1, wherein the determining the target sub-segment from the set of sub-segments that the enterprise knowledge question matches based on the enterprise knowledge question vector and the feature vector of each sub-segment in the set of sub-segments comprises: Performing attribute classification on the enterprise knowledge questioning contents to obtain attribute categories of the enterprise knowledge questioning contents; Under the condition that the attribute category of the enterprise knowledge questioning content comprises a knowledge query category, determining a target sub-segment matched with the enterprise knowledge questioning content from the sub-segments corresponding to the enterprise knowledge text under the knowledge query category according to the enterprise knowledge questioning vector and the feature vector of the sub-segment corresponding to the enterprise knowledge text under the knowledge query category; under the condition that the attribute category of the enterprise knowledge questioning content comprises a business query category, determining candidate enterprise knowledge texts which can be accessed by a target questioner of the enterprise knowledge questioning content from enterprise knowledge texts under the business query category according to the authority of the target questioner of the enterprise knowledge questioning content; And determining a target sub-segment matched with the enterprise knowledge question content from the sub-segments corresponding to the candidate enterprise knowledge texts according to the enterprise knowledge question vector and the feature vector of the sub-segment corresponding to the candidate enterprise knowledge texts.
5. The method of claim 4, wherein the determining the target sub-segment for matching the enterprise knowledge question from the sub-segments corresponding to the candidate enterprise knowledge text based on the enterprise knowledge question vector and the feature vector of the sub-segment corresponding to the candidate enterprise knowledge text comprises: Carrying out knowledge classification on the enterprise knowledge questioning content to obtain questioning knowledge categories; Determining a plurality of sub-segments to be determined, which are matched with the questioning knowledge category, according to the segment categories of the sub-segments corresponding to the candidate enterprise knowledge text, wherein the segment categories of the sub-segments corresponding to the candidate enterprise knowledge text are included in the segment categories of the sub-segments in the sub-segment set, and the segment categories of the sub-segments in the sub-segment set are obtained by classifying the knowledge of the sub-segments associated with the parent segments according to at least one knowledge category of the parent segments in the parent segment set; And determining a target sub-segment matched with the enterprise knowledge question content from the plurality of to-be-tested sub-segments matched with the question knowledge category according to the feature vectors of the plurality of to-be-tested sub-segments matched with the enterprise knowledge question vector and the question knowledge category.
6. The method of any of claims 1-5, wherein the generating and outputting enterprise knowledge answer content for the enterprise knowledge question content based on the enterprise knowledge question content, the target sub-segment, and the target parent segment comprises: Determining a target answer strategy according to the working years of the target questioner of the enterprise knowledge questioning content; Extracting target knowledge content from the target sub-segment and the target parent segment according to the target answer strategy; and generating and outputting enterprise knowledge answer content aiming at the enterprise knowledge questioning content according to the enterprise knowledge questioning content and the target knowledge content.
7. The method according to any one of claims 1-5, further comprising: Acquiring an enterprise knowledge document library, an enterprise communication data content library and an enterprise knowledge multimedia content library; converting the images in the enterprise knowledge document library into text description information to obtain a first knowledge text library; Extracting enterprise knowledge content related to enterprise knowledge from the enterprise communication data content library, and converting multimedia content in the enterprise knowledge content into text description information to obtain a second knowledge text library; converting each enterprise knowledge multimedia content in the enterprise knowledge multimedia content library into text description information to obtain a third knowledge text library; and merging the first knowledge text base, the second knowledge text base and the third knowledge text base to obtain the enterprise knowledge text base.
8. An enterprise knowledge question and answer device, characterized in that the enterprise knowledge question and answer device comprises: the vector acquisition module is used for responding to the acquired enterprise knowledge questioning content and carrying out vectorization processing on the enterprise knowledge questioning content to obtain an enterprise knowledge questioning vector; The matching module is used for determining target sub-segments matched with the enterprise knowledge question content from the sub-segment set according to the enterprise knowledge question vector and the feature vector of each sub-segment in the sub-segment set, wherein the sub-segment set comprises at least one sub-segment associated with each father segment in the father segment set, and the at least one sub-segment associated with each father segment is obtained by carrying out segment processing on each enterprise knowledge text in an enterprise knowledge text base; The parent segment determining module is used for determining a target parent segment associated with the target sub-segment from parent segments associated with all sub-segments in the sub-segment set; and the generating and outputting module is used for generating and outputting enterprise knowledge answer content aiming at the enterprise knowledge questioning content according to the enterprise knowledge questioning content, the target sub-segment and the target father segment.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.

Description

Enterprise knowledge question and answer method and device, computer equipment and storage medium Technical Field The present application relates to the field of data processing, and in particular, to an enterprise knowledge question-answering method, an enterprise knowledge question-answering device, a computer device, and a storage medium. Background Under the current acceleration of digital transformation, enterprises accumulate massive knowledge resources in the long-term operation and development process, and are important components of enterprise core competitiveness. However, how to quickly and accurately obtain the required knowledge content from huge enterprise knowledge resources has become a key problem to be solved in the field of enterprise knowledge management. In related technology, enterprise knowledge content is often obtained through an office automation system of an enterprise, for example, the office automation system queries based on a keyword matching search technology, and when a user inputs a query keyword, the office automation system performs knowledge matching in a stored knowledge base and returns the enterprise knowledge content containing the keyword. However, in the related art, the keyword-based enterprise knowledge content obtaining manner often queries a large amount of knowledge content, and then needs to screen the required knowledge content from the large amount of knowledge content, which results in the problem of low enterprise knowledge query efficiency. Disclosure of Invention Based on the method, the device, the computer equipment and the storage medium, the application provides the enterprise knowledge question-answering method, the device, the computer equipment and the storage medium, and can improve the query efficiency of enterprise knowledge. In a first aspect, the present application provides an enterprise knowledge question-answering method, which is applied to an enterprise knowledge question-answering system, and the method includes: responding to the acquired enterprise knowledge questioning content, and carrying out vectorization processing on the enterprise knowledge questioning content to obtain an enterprise knowledge questioning vector; Determining target sub-segments matched with enterprise knowledge question contents from a sub-segment set according to the enterprise knowledge question vector and feature vectors of each sub-segment in the sub-segment set, wherein the sub-segment set comprises at least one sub-segment associated with each father segment in a father segment set, and the at least one sub-segment associated with each father segment is obtained by carrying out segment processing on each enterprise knowledge text in an enterprise knowledge text library; Determining a target parent segment associated with a target sub-segment from parent segments associated with each sub-segment in the sub-segment set; And generating and outputting enterprise knowledge answer contents aiming at the enterprise knowledge questioning contents according to the enterprise knowledge questioning contents, the target sub-segments and the target parent segments. In some embodiments, determining a target sub-segment from the set of sub-segments that matches the content of the enterprise knowledge question based on the enterprise knowledge question vector and the feature vector of each sub-segment in the set of sub-segments, comprises: Determining feature vectors of a plurality of sub-segments to be selected according to the similarity between the feature vector of each sub-segment in the sub-segment set and the enterprise knowledge question vector; determining the matching degree between the feature vector of each sub-segment to be selected and the enterprise knowledge question vector according to the similarity between the feature vector of each sub-segment to be selected and the enterprise knowledge question vector and according to the validity weight of each sub-segment to be selected, wherein the validity weight of each sub-segment to be selected is included in the validity weight of each sub-segment in the sub-segment set, and the validity weight of each sub-segment in the sub-segment set is determined according to the data source of each sub-segment in the sub-segment set, the disclosure time of each sub-segment in the sub-segment set and the version of each sub-segment in the sub-segment set; and determining target subsections matched with the enterprise knowledge question content from the subsections set according to the matching degree between the feature vector of each subsection to be selected and the enterprise knowledge question vector. In some embodiments, the method further comprises: Acquiring accuracy feedback information of enterprise knowledge answer content; determining the current feedback accuracy of the target sub-segment according to the accuracy feedback information; And adjusting the effectiveness weight of the target sub-segment according