Multi-label classification(MLC) is a machine learning problem where each instance may belong to more than one class at the same time. Due to overlapping classes and label-label correlation, solving MLC is very challenging. Further, class imbalance and computational time-complexity are also considered to be major issues. In this paper, we have proposed a novel multi-label classifier that addressed the aforementioned issues; termed as Binary-Tree based Mean-Averaging estimation for Multi-label classification (BT-MA). This proposed classifier takes distinct label-sets meta-feature into account for recovering data imbalance and employs the Divide-and-conquer strategy for resolving time-complexity issue. The experimental results on several benchmark data sets show that our proposed approach BT-MA is as competitive as other Multi-label classification approaches.
https://link.springer.com/chapter/10.1007/978-3-031-78192-6_18