News Mining Using Evolving Fuzzy Systems

摘要

Online news has become one of the major channels for Internet users to get news. Modern society generates huge amounts of online newspapers every day. Thus, the processing and analysis of this information is an important challenge. In this paper, we present an approach for classifying different news articles into various topic areas based on the text content of the articles. In order to achieve this task, we need to take into account that there are thousands of new articles each day and also that articles of the same topic can vary according to the present time. For this reason, the presented approach is based on Evolving Fuzzy Systems (EFS) and the model that describes a topic area changes according to the change in the text content of the articles. This approach has been successfully tested using real on-line news.