Advertisement

Multi-label Learning

  • Sarah VluymansEmail author
Chapter
Part of the Studies in Computational Intelligence book series (SCI, volume 807)

Abstract

The defining characteristic of multi-label as opposed to single-label data is that each instance can belong to several classes at once. The multi-label classification task is to predict all relevant labels of a target instance. This chapter presents and experimentally evaluates our FRONEC method, the Fuzzy Rough NEighbourhood Consensus.

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of Applied Mathematics, Computer Science and StatisticsGhent UniversityGentBelgium

Personalised recommendations