Bullet screen sentiment analysis method based on time features

Abstract

The invention provides a time feature-based bullet screen sentiment analysis method, which comprises the following steps of: firstly, carrying out text preprocessing such as word segmentation and stop word removal on original bullet screen data to improve the embedding accuracy; and then, an embedded module in the bullet screen sentiment analysis model is used for converting words into characters and outputting fusion vectors with time characteristics, so that the problem of complicated bullet screen structure is solved, the influence of network vocabularies on semantics is reduced, the model can better learn hidden semantics of the bullet screen, and the bullet screen analysis performance of the model is optimized. Secondly, deep features of bullet screen data are further obtained through a bidirectional long-short term memory model BiLSTM and an attention mechanism, and two factors of text and time are fully considered; and finally, accurately classifying the emotion types of the bullet screens through a classifier. Compared with an existing bullet screen analysis method, the bullet screen sentiment analysis method effectively improves the accuracy of bullet screen sentiment analysis and has high applicability.

Description

technical field [0001] The invention belongs to the field of text classification, and in particular relates to a barrage sentiment analysis method based on time features. Background technique [0002] Text sentiment analysis is a kind of text classification, which refers to the process of analyzing, processing and extracting subjective texts with emotional color using natural language processing and text mining technology. It has formed an influential research field. And has been widely used in industry. The need for sentiment analysis stems from the interest of individuals, businesses, and governments in products, marketing, and policy. In recent years, the needs of innovation and business have continuously driven the development of sentiment analysis research. [0003] With the development of the Internet and social media, people's demand for entertainment consumption is getting higher and higher. According to the Statistical Report on Internet Development in China, onl...

AI Technical Summary

Problems solved by technology

However, bullet chatting is different from general microblogs, product reviews, and movie reviews. It has various structures, complex composition, and many omitted information. It is a special kind of short text. low rate

Images