Open AccessThe Research Progress of the Deep Hybrid Model in the Field of Text Classification
Author(s) -
Chunfeng Yang,
Qiang Wu,
Jiajia Lu,
Huiyu Chen
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 - 10.1088/1742-6596/2010/1/012041
Subject(s) - computer science , convolutional neural network , artificial intelligence , field (mathematics) , deep learning , artificial neural network , feature (linguistics) , feature extraction , convolution (computer science) , context (archaeology) , pattern recognition (psychology) , machine learning , data mining , paleontology , linguistics , philosophy , mathematics , pure mathematics , biology
The traditional deep convolution neural network model cannot extract the context information effectively in dealing with complex long text data sets, and it is difficult to obtain the deep semantic information of the text. The deep hybrid neural network can optimize and improve the local feature extraction ability of CNN model while preserving the ability of local feature extraction. It has achieved good performance on complex data sets, so it has been paid more and more attention by researchers. Firstly, this article sorts out the current mainstream text classification data sets. Secondly, the model of the hybrid neural network based on the convolutional neural network construction is as follows: the improvement of the CNN model; technology fusion based on CNN model and the CNN-based model mixing three categories carry out analysis and sorting. Finally summarize the current problems in the text classification field, and look forward to future development and research.