UofAComputing ScienceSemester 2006-3

Mining Web Links
(Independent Study)
Instructor: Osmar R. Zaļane

OBJECTIVE/DESCRIPTION:

One of the emerging data mining challenges the discovery of interesting patterns in complex datasets with interconections such as links and hyperlinks in data such as web content and social networks. Many other real datasets are described and contain a variety of entity types connected by means of relationships.

This course will provide the students with (1) an overview the current state of the mining techniques dedicated to such data, (2) an opportunity to thoroughly study and compare existing techniques to mine linked data (web structure data and social networks) and compare objectively, (3) analyse visualization tools and approaches dedicated linked data, and (3) an opportunity to critically discuss the shortcomings of existing methods and design an appropriate approach for web stucture and web usage analysis and pattern visualization in link data context. The students will be provided with enough background so that a term project (prototype) can be developed.

The course will mainly consist of a series of discussions on the topics listed below as a general guideline. Throughout the course recent relevant research papers will also be read/discussed, particularly the 10 research papers published in the special issue of the SIGKDD Exploration on Link Mining, Volume 7, Issue 2.

TOPICS:

The course will cover the following topics:

GRADING:

Annotated Bibliography (20%),
Discussions (20%),
Implementation and testing (20%)
Final Term paper (40%).

TEXTBOOK and REFERENCES:


Distributed: August, 2006