Text retrieval method and device

Abstract

The embodiment of the invention provides a text retrieval method and device. The text retrieval method includes the steps that an original text input by a user is acquired; retrieval words are acquired from the original text; according to the retrieval requirement of the user, the retrieval words are filtered to acquire keywords; the keywords are combined, texts in a text database are retrieved according to the combined keywords, and at least one retrieval text is acquired; the retrieval texts are displayed in a relevancy inverted order mode, and the keywords are highlighted in the retrieval texts, wherein relevancy is used for representing the relevancy degree of the original text and the retrieval texts. Due to the fact that the keywords are acquired by filtering the retrieval words according to the retrieval requirement of the user, the probability that the keywords are invalid words is reduced, and the retrieval requirement is better met compared with the manner that the retrieval words are directly acquired from the original text, the retrieval texts acquired through retrieval by the application of the combined keywords can well meet the retrieval requirement, and therefore retrieval accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of text mining, in particular to a text retrieval method and device. Background technique [0002] As the name suggests, text retrieval is to extract valuable information from text, and display the valuable information to the user on the display screen of the electronic device. At present, the principle of text retrieval is to perform matching search directly after text segmentation. There are two common text retrieval methods: providing users with a search method of classified browsing or a full-text search method based on keywords. [0003] The above two retrieval methods obtain keyword matching by segmenting the text input by the user, but the keywords obtained by this simple text segmentation method include invalid words, wherein the invalid words appear in most texts and cause Search for words with increased results. For example, the text entered by the user is: an image matching device based on an im...

AI Technical Summary

Problems solved by technology

[0004] From the above technical solutions, it can be seen that the existing text segmentation cannot fully understand the user's retrieval needs, so the keywords extracted during text segmentation may be invalid words, and keywords that are not invalid words may not meet the user's retrieval requirements requirements, and then in the fuzzy full-text retrieval based on these keywords, there will be texts that do not meet the retrieval requirements in the retrieval results, and the retrieval accuracy will be reduced

Images