University of Konstanz
Graduiertenkolleg / PhD Program
Computer and Information Science

Graduation Talks

title

Multi-label Classification with Multiple Data Sources and Taxonomies

speaker

Fernando Benites, University Konstanz
Konstanz, Germany

date & place

Wednesday, 09.02.2011, 16:15 h
Room C 252

abstract

In the field of data mining several methods have been developed to investigate the increasing complexity in classification problems both in the input space and in the output space,including such methods as Multi-View Learning and Multi-Label Classification(MC). This work will examine the combination of complex structured input spaces and complex structured output spaces, focusing on the problems emerging from hierarchical MC with multiple heterogeneous sources (HMC-MHS), examining in particular taxonomies, that are very different in nature, and sources, which have the same property. Another objective of this work is to develop an interactive rule exploration system.