TY - CONF ID - jf:PL-09-WS-Paper T1 - On the Combination of Two Decompositive Multi-Label Classification Methods A1 - Tsoumakas, Grigorios A1 - Loza Mencía, Eneldo A1 - Katakis, Ioannis A1 - Park, Sang-Hyeun A1 - Fürnkranz, Johannes ED - Hüllermeier, Eyke ED - Fürnkranz, Johannes TI - Proceedings of the ECML PKDD 2009 Workshop on Preference Learning (PL-09, Bled, Slovenia) Y1 - 2009 SP - 114 EP - 129 UR - /events/PL-09/09-Tsoumakas.pdf N2 - 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. ER -