Automatic Recognition of News Web Pages

摘要

The information on the World Wide Web is congested with large amounts of news contents. The filtering, summarization and classification of news Web pages have become hot topics of research, aiming for useful news contents. Accurately identifying news Web pages is a crucial problem in these research topics. To solve this problem, this paper proposes an automatic recognition method for news Web pages based on a combination of URL attributes, structure attributes and content attributes. Our experimental results demonstrate that this method provides a high accuracy of above 96% with the recognition of news Web page.