Fine-grained sentiment analysis method based on position enhancement

Abstract

The invention provides a fine-grained sentiment analysis method based on position enhancement, which is used for solving the problem of low precision caused by performing fine-grained sentiment analysis on a text in the prior art. Firstly, a text is preprocessed, and then sentiment analysis is carried out through a fine-grained sentiment analysis model. The model comprises an embedding layer, a semantic representation layer, an information interaction layer and an output layer. The embedding layer is used for mapping sentences into context word embedding and aspect word embedding, the semantic representation layer is used for enhancing the text semantic representation capability of a model through a position intensified attention mechanism, and the information interaction layer is used for enhancing the interactivity between aspect words and contexts of the aspect words by using a memory network; and the aspect-based context semantic enhancement representation is used as an external memory unit interacting with the aspect, so that the external memory unit can learn semantic information in a complex text, and finally an output layer performs emotion prediction. According to the method, the context range for performing emotion expression on the aspect is reasonably defined, so that the fine-grained emotion analysis accuracy is improved.

Description

technical field [0001] The invention belongs to the technical field of information processing, and is a fine-grained emotion analysis method based on location enhancement. Background technique [0002] The rapid development of social networks and e-commerce shopping platforms makes it easier for people to express their views and opinions on the network platform, thus generating a large amount of text data containing user emotional information, which contains huge practical value. And products have multi-dimensional attributes, and consumers will comment on products and services from different angles, such as quality, price and service. The traditional text sentiment analysis technology usually gives an emotional judgment as a whole, which cannot meet the needs of different aspects of the emotional orientation judgment in the comment text, so the granularity of text sentiment analysis needs to be more refined. For example, in the sentence "The food in this restaurant is deli...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem of low accuracy caused by fine-grained sentiment analysis of text in the prior art, the present invention proposes a fine-grained sentiment analysis method based on location enhancement

Images