Using HTML Formatting to Aid in Natural Language Processing on the World Wide Web

Because of its magnitude and the fact that it is not computer understandable, the WorldWide Web has become a prime candidate for automatic natural language tasks. This thesisargues that there is information in the layout of a web page, and that by looking at theHTML formatting in addition to the text on a page, one can improve performance in taskssuch as learning to classify segments of documents. A rich representation for web pages, theHTML Struct Tree, is described. A parsing algorithm...