Sentiment Analysis Method of Network Text Based on Improved AT-BiGRU Model

• Published in: Scientific programming Vol. 2021; pp. 1 - 11
• Main Authors: Lu, Xinxin, Zhang, Hong
• Format: Journal Article
• Language: English
• Published: New York Hindawi 2021; John Wiley & Sons, Inc
• Subjects: Accuracy; Classification; Cost analysis; Data mining; Deep learning; Dictionaries; Emotions; Iterative methods; Machine learning; Model accuracy; Natural language processing; Network analysis; Neural networks; Research methodology; Sentiment analysis; Spelling

In order to solve the problems existing in the current method of emotional analysis of network text, such as long training time, complex calculation, and large space cost, this paper proposes an Internet text sentiment analysis method based on the improved AT-BiGRU model. Firstly, the textblob packa...