Web Text Categorization Based on Statistical Merging Algorithm in Big Data Environment
Rujuan Wang, Gang Wang · International Journal of Ambient Computing and Intelligence · 2019
In the field of modern information technology, how to find information quickly, accurately and comprehensively that users really needed has become the focus of research in this field. In this article, a feature selection method based on a complex network is proposed for the structure and content characteristics of large-scale web text information. The preprocessed web text is converted into a complex network. The nodes in the network correspond to the entries in the text. The edges of the network correspond to the links between the entries in the text, and the degree of nodes and the aggregation system are used. Second, the text classification method is studied from the point of view of data sampling, and a text classification method based on density statistics is proposed. This method uses not only the density information of the text feature set in the classification process, but also the use of statistical merging criteria to get the text. The difference information of each feature has a better classification effect for large text collections.