Theseus: Categorization by Context
Giuseppe M. Attardi, Antonio Gullì · 2000
Introduction The traditional approach to document categorization is categorization by content, since information for categorizing a document is extracted from the document itself. In a hypertext environment like the Web, the structure of documents and the link topology can be exploited to perform what we call categorization by context [Attardi 98]: the context surrounding a link in an HTML document is used for categorizing the document referred by the link. Categorization by context is capable of dealing also with multimedia material, since it does not rely on the ability to analyze the content of documents. Categorization by context leverages on the categorization activity implicitly performed when someone places or refers to a document on the Web. By focusing the analysis to the documents used by a group of people, one can build a catalogue tuned to the need of that group. Categorization by context is based on the following assumptions: 1