Refereed Papers Track
The web has become the primary repository of the world's
knowledge. Billions of pages, maintained by individuals and organizations
around the world contain content, services and semantic associations on a
wide range of topics. There are also many additional (possibly implicit)
corpora associated with the web such as hyperlink usage data, site
transaction logs, message boards, query streams, and chat sessions. Thus
the web provides a rich domain for using data mining techniques to both
extract useful knowledge from this huge and diverse corpus, and to study
the structural properties (size, shape, growth) and social aspects of the
web (hubs, communities, weblogs).
The focus of the Data Mining area of the WWW 2004 refereed papers
track includes, but is not limited to, the following topics:
- Machine learning and mining for unstructured,
- Classification, clustering, collaborative recommender
- Adapting data mining techniques for web data.
- Deriving facts, associations and lists from web
- Information extraction and message understanding.
- Mining associated corpora: queries, usage, messages,
- Mining using hyperlink graph analysis, e.g., communities,
structural properties, evolution.