Text processing method and device and storage medium

Abstract

The invention provides a text processing method. The method comprises the steps that according to an existing user comment text of a target domain, acquiring an emotion matching dictionary of the target domain, wherein the emotion matching dictionary comprises a plurality of target emotion matching groups, and each target emotion matching group is used for representing emotion comments of attributes of a comment object of the target domain by a user; and according to the to-be-processed user comment text and the emotion matching dictionary in the target field, obtaining an emotion matching group corresponding to the to-be-processed user comment text. According to the text processing method provided by the invention, the sentiment matching dictionary of the target domain is constructed according to the existing user comment text, and then the sentiment matching dictionary is adopted to obtain the sentiment matching group of the to-be-processed text, so that the sentiment viewpoint of the text of the target domain can be accurately obtained.

Description

technical field [0001] The present invention relates to the technical field of text sentiment analysis, in particular to a text processing method, device and storage medium. Background technique [0002] User comment text (UGC text) is the text that the user evaluates the product after experiencing the product, and the extraction of emotional views in the text is very important; for example, the automatic comment text is the comment text of a restaurant, including "the taste of this restaurant Not bad", "I really like the taste here" and "This beef soup is delicious", the extraction result of the emotional point of view of the above comment text is "taste, good, positive"; the extraction result of the emotional point of view in the text It can enable merchants to see users' impressions of their own products, and optimize products in a more targeted manner. It also enables users to make better consumption decisions by comparing reviews from different merchants. [0003] In t...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, it is possible to realize the extraction of emotional viewpoints in the general technical field, but because the attribute words of the emotional viewpoints in the general technical field are applicable to all fields, the attribute words of the emotional viewpoints are too single and are not suitable for vertical fields; For example, the extraction result of the emotional viewpoint in the general technology field is "design, good, positive", which is not applicable to the comment text in the food field "this restaurant tastes good"

Images