On the Combination of Two Decompositive Multi-Label Classification Methods
Type of publication: | Inproceedings |
Citation: | jf:PL-09-WS-Paper |
Booktitle: | Proceedings of the ECML PKDD 2009 Workshop on Preference Learning (PL-09, Bled, Slovenia) |
Year: | 2009 |
Pages: | 114--129 |
URL: | /events/PL-09/09-Tsoumakas.pdf |
Abstract: | In this paper, we compare and combine two approaches for multi-label classication that both decompose the initial problem into sets of smaller problems. The Calibrated Label Ranking approach is based on interpreting the multi-label problem as a preference learning problem and decomposes it into a quadratic number of binary classiers. The HOMER approach reduces the original problem into a hierarchy of considerably simpler multi-label problems. Experimental results indicate that the use of HOMER is benecial for the pairwise preference-based approach in terms of computational cost and quality of prediction. |
Keywords: | |
Authors | |
Editors | |
Attachments
|
|
Topics
|
|
|