Key term extraction method, device and equipment and computer readable storage medium

Abstract

The invention discloses a key term extraction method, device and equipment and a computer readable storage medium, and the method comprises the steps of carrying out the segmentation of a text according to a pre-constructed specific domain term dictionary; traversing the text by using a preset first extraction window, extracting the words obtained after segmentation processing to obtain candidateterms of a specific field, and extracting words obtained after segmentation processing according to a pre-constructed term dictionary of the specific field to obtain candidate terms of the specific field; performing the topic clustering on the candidate terms through a pre-constructed probability topic model to obtain a plurality of topic-associated candidate terms and association probabilities thereof; and determining the key terms according to the candidate terms associated with each topic and the association probability of the candidate terms. According to the present invention, the text isdivided on the basis of the term dictionary in the specific field, and the probability topic model is adopted for extracting the key terms, so that the key terms in the specific field are effectivelyextracted, and the key term extraction accuracy is improved.

Description

technical field [0001] The present invention relates to the technical field of word extraction, in particular to a key term extraction method, device, equipment and computer-readable storage medium. Background technique [0002] The traditional Chinese key term extraction method is generally to first segment the Chinese text, and then based on the word segmentation results, the frequency method is used to extract key terms. However, the traditional Chinese key term extraction method has the following defects: (1) Using a general segmentation dictionary, it is easy to segment the key terms in a specific field into words that are not related to the subject of the article, resulting in the final key term extraction cannot express the real meaning of the article. (2) Pay too much attention to the frequency of words, ignoring some low-frequency key terms. Contents of the invention [0003] In view of the above problems, the object of the present invention is to provide a key t...

AI Technical Summary

Problems solved by technology

However, the traditional Chinese key term extraction method has the following defects: (1) Using a general segmentation dictionary, it is easy to segment the key terms in a specific field into words that are not related to the subject of the article, resulting in the final key term extraction cannot express the real meaning of the article. (2) Pay too much attention to the frequency of words, ignoring some low-frequency key terms

Images