A library intelligent search sorting method and system

Abstract

The invention discloses an intelligent search sorting method of a library. The intelligent search sorting method comprises the following steps: firstly, establishing an intelligent search model training device and an intelligent search predication sorting model; obtaining search information; treating the search information by utilizing a cyclic convolution network and extracting an information characteristic vector; carrying out normalization computing on the characteristic vector of the search information by utilizing a softmax function; calculating a book sequence through the intelligent search predication sorting model and storing in a database; predicating an error degree by utilizing an accumulated error computing model; determining whether the intelligent search predication sorting model is updated or not according to the size of an accumulated error value. Compared with a traditional search system matched with a title and an author keyword, the method provided by the invention is more flexible and the search accuracy of a searched keyword which is not in the book title is improved; the intelligent search model training device and the database are arranged so that the system has a self-adaptive learning capability and manpower participation is reduced; manpower resources are saved and the utilization difficulty are saved; an intelligent search sorting system which has own book characteristics of the library and habits of a lot of users can be more conveniently distributed.

Description

technical field [0001] The invention relates to the technical field of automatic book detection, in particular to a library intelligent search and sorting method and system for accurate prediction and recommendation. Background technique [0002] Existing library book retrieval methods generally use keyword matching. At the same time, in order to reduce workload and complexity, most retrieval systems use keyword matching of book titles and authors. The search for keywords that do not match the title of the book will fail; [0003] Therefore, there is a need for an intelligent search and ranking method and system for libraries that can accurately predict and recommend. Contents of the invention [0004] The object of the present invention is to propose a method and system for intelligently searching and sorting libraries for accurate prediction and recommendation. [0005] The purpose of the present invention is achieved through the following technical solutions: [0006...

AI Technical Summary

Problems solved by technology

[0002] Existing library book retrieval methods generally use keyword matching. At the same time, in order to reduce workload and complexity, most retrieval systems use keyword matching of book titles and authors. The search for keywords that do not match the title of the book will fail;

Images