TY - GEN
T1 - Multi-label classification using ensembles of pruned sets
AU - Read, Jesse
AU - Pfahringer, Bernhard
AU - Holmes, Geoff
PY - 2008/12/1
Y1 - 2008/12/1
N2 - This paper presents a Pruned Sets method (PS) for multilabel classification. It is centred on the concept of treating sets of labels as single labels. This allows the classification\ process to inherently take into account correlations between labels. By pruning these sets, PS focuses only on the most important correlations, which reduces complexity and improves accuracy. By combining pruned sets in an ensemble scheme (EPS), new label sets can be formed to adapt to irregular or complex data. The results from experimental evaluation on a variety of multi-label datasets show that [E]PS can achieve better performance and train much faster than other multi-label methods.
AB - This paper presents a Pruned Sets method (PS) for multilabel classification. It is centred on the concept of treating sets of labels as single labels. This allows the classification\ process to inherently take into account correlations between labels. By pruning these sets, PS focuses only on the most important correlations, which reduces complexity and improves accuracy. By combining pruned sets in an ensemble scheme (EPS), new label sets can be formed to adapt to irregular or complex data. The results from experimental evaluation on a variety of multi-label datasets show that [E]PS can achieve better performance and train much faster than other multi-label methods.
UR - https://www.scopus.com/pages/publications/67049088703
U2 - 10.1109/ICDM.2008.74
DO - 10.1109/ICDM.2008.74
M3 - Conference contribution
AN - SCOPUS:67049088703
SN - 9780769535029
T3 - Proceedings - IEEE International Conference on Data Mining, ICDM
SP - 995
EP - 1000
BT - Proceedings - 8th IEEE International Conference on Data Mining, ICDM 2008
T2 - 8th IEEE International Conference on Data Mining, ICDM 2008
Y2 - 15 December 2008 through 19 December 2008
ER -