Open AccessA Text classification algorithm based on topic model and convolutional neural network
Author(s) -
Junwei Ge,
Songce Lin,
Yiqiu Fang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1748/3/032036
Subject(s) - computer science , convolutional neural network , artificial intelligence , artificial neural network , feature (linguistics) , granularity , layer (electronics) , process (computing) , pattern recognition (psychology) , philosophy , linguistics , chemistry , organic chemistry , operating system
Based on the neural topic model ProdLDA and convolutional neural network, this paper proposes a text classification algorithm based on topic model and convolutional neural network. Firstly, the text information is modeled on the word vector model, then the convolutional neural network is used to extract the granularity features of high-dimensional text, and the neural topic model ProdLDA is used to extract the potential topic features. Then, the connection layer is established to connect the text features, and finally the classification layer is processed. At the same time, a new topic feature introduction method is used in the process of extracting topic features. Experimental results show that this algorithm can effectively improve the performance of text classification.