A Survey on Social Web Mining and Related Data Mining Tools

PhD Qualifying Examination


Title: "A Survey on Social Web Mining and Related Data Mining Tools"

Mr. Nan Liu


Abstract:

Social Web (or Web 2.0) applications have rapidly emerged on the Web. This 
indicates a currently ongoing grass-root creation of knowledge spaces on the 
Web. The reason for the success of the Social Web lies mainly in the fact that 
there is very low barrier of entry for publishing and editing while the 
self-organized collaboration among massive users autonomously ensure the high 
quality and broad coverage in the content. Web 2.0 applications are a very 
interesting application area for data mining. Unlike in traditional data mining 
scenarios, data does not emerge from a small number of (heterogeneous) data 
sources, but virtually from millions of different sources. As there is only 
minimal coordination, these sources can overlap or diverge in any possible way. 
This fundamental structure, known from collaborative filtering, is not limited 
to ratings and recommendations but can be applied to arbitrary complex data and 
data mining tasks. In this survey, we review existing data mining research 
related to the two most common forms of social Web application, namely 
collaborative content creation (eg. Wikipedia) and collaborative resource 
sharing (e.g. Delicious, Flickr). In addition, we also survey collaborative 
filtering and multi-label classification, which are two areas within machine 
learning and data mining that are highly useful for modeling data on the social 
web.


Date:     		Thursday, 11 June 2009

Time:                   11:00am-1:00pm

Venue:                  Room 3501
 			lifts 25-26

Committee Members:      Prof. Qiang Yang (Supervisor)
 			Dr. Raymond Wong (Chairperson)
 			Prof. Dik-Lun Lee
 			Prof. Dit-Yan Yeung


**** ALL are Welcome ****