Research On Network News Text Classification Based On Deep Learning

As an important means for people to obtain effective social information in the era of big data,online news has received widespread attention.It realizes the intel igent and efficient classification of massive network news,and has important significance for personalized news recommendation,topic identification and tracking,and news website classification and navigation.With the development of computer technology and the popularity of the Internet,online news information resources have exploded.Faced with massive and chaotic network news information,the problem faced by users is no longer how to obtain news information,but how to quickly and effectively find information that can meet their own needs in large-scale news resources.With the development of the information age,users' requirements for the content and quality of information are also increasing.News information is an important channel for online news users to obtain social information resources.With the development and maturity of text classification technology,it lays the foundation for the application and practice of text categorization in the field of online journalism,and the development of online news text classification has ushered in a new opportunity.In order to improve the quality of information services in the field of online news and meet the diverse and personalized information needs of users in the era of big data.This paper is based on an in-depth study of the background,research status,relevant theories and development of online news text classification.This paper uses the bibliometric method to statistica l ly analyze the relevant literature published in the field of text classification in the past fifteen years,and clarify the field of text classification from the aspects of annual publicat io n trends,disciplinary distribution,journal distribution,institutional distribution,author distribution,research hotspots and frontiers.The development context and research status provide a theoretical reference for promoting the further development of text classification.At the same time,this paper builds an efficient network news text classification model framework for the network news field.The model is mainly composed of four parts,namely news text preprocessing,news text representation based on word vector,news text feature extraction and classification,and text classifica t io n result evaluation.In the process of experiment,the dense Word2 Vec word vector representation of news text is firstly learned,which solves the problems of high latitude,data sparseness and lack of semantics of traditional text representation,and then uses the news text word vector as input and convolves through neural network.The characterist ics of news texts are automatically learned and extracted,which avoids the disadvantages of time-consuming and labor-intensive extraction of traditional news text classifica t io n methods.The experimental results show that this method can effectively improve the efficiency of online news text classification and promote the realization of news field.Effective information organization and management.The network news text classification model based on deep learning proposed in this paper is feasible,can provide users with better news information services,and provides certain reference value for the development of network news text classification technology.