Feature Selection And Feature Weight Estimate In Web Text Mining

Text mining is a combination of data mining technology and text. It refers to the extraction of interest or knowledge or information from the data file on the server. The most common is web text mining. Its essence is that The contents of the document, available resources and the relationships between resources are analyzed, and what we want is to be found. This paper designs and implements an algorithm of hierarchical processing. When the characteristics of the original text are extracted and classified, a level of importance is given, so that the system can automatically perform feature selection and feature weight estimation. Experiments show that this new feature weighting method and hierarchical processing algorithm is correct and feasible.