Open AccessNews text classification based on hybrid model of Bidirectional Encoder Representation from Transformers and Convolutional Neural Network
Author(s) -
Wei Fan,
Fan Li
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 - 10.1088/1742-6596/2005/1/012021
Subject(s) - computer science , encoder , convolutional neural network , rss , transformer , embedding , information retrieval , the internet , artificial intelligence , data mining , world wide web , physics , quantum mechanics , voltage , operating system
At present, with the rapid development of Internet technology, text data show massive characteristics. Network media and new media platforms have become an important part of journalism and communication, and network news has become one of the important sources of information. In order to meet the needs of online news reading users and improve the efficiency of personalized news recommendation by content distribution platform, it is urgent to effectively manage and utilize online news. Based on this, this paper studies and implements the algorithm for news text classification, and the goal is to build a classification model for news text with higher classification accuracy. In this paper, BERT and CNN algorithm are combined to classify news texts. The core idea is to send BERT as the embedding layer into the CNN model. The experimental results show that this method is better than the simple BERT model or the simple CNN model.