Fuzzy ARTMAP with binary relevance for multi-label classification

In this paper, we propose a modified supervised adaptive resonance theory neural network, namely Fuzzy ARTMAP (FAM), to undertake multi-label data classification tasks. FAM is integrated with the binary relevance (BR) technique to form BR-FAM. The effectiveness of BR-FAM is evaluated using two bench...

Full description

Main Authors: Yuan, L.X., Tan, S.C., Goh, P.Y., Lim, C.P., Watada, J.
Format: Article
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-eprints.21373 /
Published: Springer Science and Business Media Deutschland GmbH 2018
Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020416745&doi=10.1007%2f978-3-319-59424-8_12&partnerID=40&md5=ad2af4eaccfb6c1c8fc243100d211971
http://eprints.utp.edu.my/21373/
Tags: Add Tag
No Tags, Be the first to tag this record!
Summary: In this paper, we propose a modified supervised adaptive resonance theory neural network, namely Fuzzy ARTMAP (FAM), to undertake multi-label data classification tasks. FAM is integrated with the binary relevance (BR) technique to form BR-FAM. The effectiveness of BR-FAM is evaluated using two benchmark multi-label data classification problems. Its results are compared with those other methods in the literature. The performance of BR-FAM is encouraging, which indicate the potential of FAM-based models for handling multi-label data classification tasks. © Springer International Publishing AG 2018.