A Survey on Multi Label Classification

Abstract

MultiLabel Classification is a kind of supervised learning where each instance can belong to set of multiple classes. The methods can be broadly classified into two groups: Problem Transformation and Algorithm Adaptation. Ten representative Multi Label models are scrutinized under common notations followed by different evaluation metrics. There exists wide range of application for multi label prediction such as text categorization, semantic image labeling, gene functionality etc. This survey paper introduces the task of multi-label classification, presents the sparse literature in this area, discusses different evaluation metrics and performs a comparative analysis of the algorithms in different tasks and application domain.

Keywords: Multi label classification, algorithm adaptation, problem transformation method, evaluation metric, approaches to MLC, supervised learning

Cite this Article
Dhatri Ganda, Rachana Buch. A Survey on Multi Label Classification. Recent Trends in Programming Languages. 2018; 5(1): 19–23p.