Text subject indexing method and device, electronic device and computer storage medium

Abstract

The embodiment of the invention relates to the technical field of text processing, and discloses a text subject indexing method and device, an electronic device and a computer storage medium, and thetext subject indexing method comprises the steps: determining a text word list of a to-be-indexed text; determining a text representation vector of the to-be-indexed text based on a predetermined wordvector library according to the text word list; then, based on a mapping table between subject words and common words, which is pre-established according to the controlled word table, determining thesubject words of which the association strength with any text word is greater than a first preset threshold value as the subject words of any text word to obtain the subject words corresponding to the text words respectively; determining a target subject word of the to-be-indexed text according to the text representation vector and the subject word corresponding to each text word, and performingsubject indexing on the to-be-indexed text through the target subject word. Therefore, the operand is greatly reduced, the comparison frequency is effectively reduced, and the text topic indexing efficiency is greatly improved.

Description

technical field [0001] The embodiments of the present application relate to the technical field of text processing, and specifically, the present application relates to a text subject indexing method, device, electronic equipment, and computer storage medium. Background technique [0002] Automatic subject indexing generally refers to the process of using a computer system to analyze, discover and extract key words used to reveal the content of documents from various elements of document composition, such as titles, keywords, abstracts and texts. Indexed documents can be expanded from papers to other forms of electronic documents such as patents, books, and web page texts. Subject indexing can be done manually or by machine. [0003] At present, automatic subject indexing methods can be divided into three categories according to technology: statistical indexing, linguistic analysis indexing and machine learning indexing. The main idea of ​​statistical indexing is that the ...

AI Technical Summary

Problems solved by technology

[0004] However, during the implementation process, the inventors of the present application found that: the accuracy rate of the statistical indexing method is low, and the selected indexing words cannot represent the content of the article well; The results are directly affected by the performance of the "rule base". Due to the complexity and flexibility of Chinese, the pre-defined rules often do not have a high degree of coverage and require a lot of manual intervention. Synonym recognition, word meaning disambiguation, etc., lead to poor versatility and greater difficulty in application; automatic indexing methods based on machine learning need to train multiple classifiers for different types of data, the training time is long, and there are problems of data sparseness and overtime. Fitting learning problems, unable to adapt to the labeling of large-scale controlled vocabularies

Images