Method for constructing fine-grained emotion classification model based on interactive attention mechanism

Abstract

The invention discloses a method for constructing a fine-grained emotion classification model based on an interactive attention mechanism, which comprises the following steps of: independently modeling semantics for an evaluation aspect and a context respectively, and simultaneously carrying out interactive learning on the semantics of the two parts by combining the attention mechanism, so as to obtain more appropriate semantic representation for fine-grained emotion classification. According to the model provided by the invention, both semantic features of a given evaluation aspect and semantic features of a given context are considered, and the two parts of features are fused, so that semantic information which is most important for sentiment analysis can be mined, and the accuracy of fine-grained sentiment classification is improved; the user emotion contained in the online comment data can be accurately understood, merchants are helped to understand the requirements of consumers, then effective decision support is provided for the consumers, and the merchants are helped to improve existing products and services.

Description

technical field [0001] The invention belongs to the technical field of computer applications, and in particular relates to a method for constructing a fine-grained emotion classification model based on an interactive attention mechanism. Background technique [0002] With the development of information technology, a large amount of user comment data has been accumulated in the Internet. Semantic analysis and opinion mining of these comment data is of great significance to the development of various industries. For example, analyzing product evaluation data in e-commerce websites can help merchants understand consumer needs and provide effective decision support for them. Help it improve existing products and services. Since the online comment data has the characteristics of wide coverage, strong colloquialism, and high noise, how to accurately understand the user emotion contained in the online comment data is the core task of the present invention. In addition, considerin...

AI Technical Summary

Problems solved by technology

However, these methods usually only use contextual semantics for sentiment classification, without considering the direct impact of evaluation semantics on sentiment classification.

Images