Towards the Learning of Weighted Multi-label Associative Classifiers
Lists
Only you see this
Abstract
Because of the ability to capture the correlation between features and labels, association rules have been applied to multi-label classification. However, existing multi-label associative classification algorithms usually exploit association rules using heuristic strategies. Moreover, only the covering association rules whose feature set is a subset of the testing instance are considered. Discarding any mined rules may diminish the performance of the classifier, especially when some rules only differ from the testing instance by a few insignificant features. In this paper we propose Weighted Multi-label Associative Classifiers (WMAC) that leverage an extended set of association rules with overlapping features with the testing instance to learn a universal weight vector for features. For this purpose, we embed the set of rules into a linear model and weigh the association rules by its confidence. Empirical results on diversified datasets clearly demonstrate that WMAC outperforms other well-established multi-label classification algorithms.
BibTeX
@inproceedings{liu2018towards,
author = {Chunyang Liu and Ling Chen and Ivor W. Tsang and Hongzhi Yin},
title = {{Towards the Learning of Weighted Multi-label Associative Classifiers}},
booktitle = {IJCNN},
year = {2018},
doi = {10.1109/ijcnn.2018.8489398},
}Loading related content...