A Book Retrieval Method Based on Deep Learning and Quality Impact

Abstract

The present invention relates to a book retrieval method based on deep learning and quality influence, comprising the following steps: 1) Obtaining data in a digital library database and extracting standardized book title data and content keyword sentence data therein, using a category noise clipping algorithm The extracted data is subjected to noise reduction processing; 2), based on the deep learning model, the data after the noise reduction processing in the above step 1) is subjected to feature extraction and classification, and the semantic class information of the book is obtained; 3), based on the acquisition in step 2) Semantic class information to build a semantic understanding index database. This book retrieval method based on deep learning and quality impact can better understand the real retrieval needs of users, and uses deep learning technology to improve the refinement of search, and introduces a book quality recognition model on the basis of ensuring the search semantic similarity , so that the semantic similarity of the results received by the user when searching the engine in the library is high, and the quality of the book is also guaranteed.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, and in particular relates to a book retrieval method based on deep learning and quality influence. Background technique [0002] Traditional library search engines lack the ability to understand natural language and do not conform to people's habitual thinking about retrieval problems. For the retrieval fields entered by users, only mechanical keyword matching can be used. [0003] Moreover, because literary works usually contain human emotions and are particular, traditional keyword extraction algorithms cannot accurately obtain truly meaningful article features. At the same time, in traditional library search engines, only the similarity ranking is considered for user results Book quality is not considered, resulting in less refined search results. Contents of the invention [0004] The purpose of the present invention is just to provide a kind of simple in structure in o...

AI Technical Summary

Problems solved by technology

[0002] Traditional library search engines lack the ability to understand natural language and do not conform to people's habitual thinking of retrieval problems. For the search fields entered by users, only mechanical keyword matching can be used.

[0003] Moreover, because literary works usually contain human emotions and are particular, traditional keyword extraction algorithms cannot accurately obtain truly meaningful article features. At the same time, in traditional library search engines, only the similarity ranking is considered for user results Doesn't take book quality into account, resulting in less granular search results

Images