News Text Classification and Recommendation Technology Based on Wide & Deep-Bert Model
Wu Jing, Yang Bailong · 2021
With the rapid development of Internet technology, news information on various social platforms is growing wildly, generating large amounts of data. Since short text information such as news headlines, short messages, and newsletters has a small number of words and limited content, it is often difficult to extract effective information. The sparse feature information causes difficulties in text classification. If the news system cannot efficiently and accurately realize news classification and users preference recommendation, it will inevitably affect the experience and frequency of platform users. This paper mainly studies the application of deep learning in the field of text classification and users' personal recommendation. It uses multiple English text data sets to learn text features. Based on the Wide&Deep model, combined with the improved BERT pretraining model, the Wide&Deep-BERT model is designed. In addition, the corresponding news text classification and recommendation technology process is proposed, and the Tensorflow deep learning framework is used to experimentally verify the technology, which proves the effectiveness and practicability of the design technology.