TY - JOUR
T1 - Hierarchical partitioning of the output space in multi-label data
AU - Papanikolaou, Yannis
AU - Tsoumakas, Grigorios
AU - Katakis, Ioannis
PY - 2018/7/1
Y1 - 2018/7/1
N2 - Hierarchy Of Multi-label classifiERs (HOMER) is a multi-label learning algorithm that breaks the initial learning task to several, easier sub-tasks by first constructing a hierarchy of labels from a given label set and secondly employing a given base multi-label classifier (MLC) to the resulting sub-problems. The primary goal is to effectively address class imbalance and scalability issues that often arise in real-world multi-label classification problems. In this work, we present the general setup for a HOMER model and a simple extension of the algorithm that is suited for MLCs that output rankings. Furthermore, we provide a detailed analysis of the properties of the algorithm, both from an aspect of effectiveness and computational complexity. A secondary contribution involves the presentation of a balanced variant of the k means algorithm, which serves in the first step of the label hierarchy construction. We conduct extensive experiments on six real-world data sets, studying empirically HOMER's parameters and providing examples of instantiations of the algorithm with different clustering approaches and MLCs, The empirical results demonstrate a significant improvement over the given base MLC.
AB - Hierarchy Of Multi-label classifiERs (HOMER) is a multi-label learning algorithm that breaks the initial learning task to several, easier sub-tasks by first constructing a hierarchy of labels from a given label set and secondly employing a given base multi-label classifier (MLC) to the resulting sub-problems. The primary goal is to effectively address class imbalance and scalability issues that often arise in real-world multi-label classification problems. In this work, we present the general setup for a HOMER model and a simple extension of the algorithm that is suited for MLCs that output rankings. Furthermore, we provide a detailed analysis of the properties of the algorithm, both from an aspect of effectiveness and computational complexity. A secondary contribution involves the presentation of a balanced variant of the k means algorithm, which serves in the first step of the label hierarchy construction. We conduct extensive experiments on six real-world data sets, studying empirically HOMER's parameters and providing examples of instantiations of the algorithm with different clustering approaches and MLCs, The empirical results demonstrate a significant improvement over the given base MLC.
KW - Knowledge discovery
KW - Machine learning
KW - Supervised learning
KW - Text mining
UR - http://www.scopus.com/inward/record.url?scp=85046851074&partnerID=8YFLogxK
U2 - 10.1016/j.datak.2018.05.003
DO - 10.1016/j.datak.2018.05.003
M3 - Article
AN - SCOPUS:85046851074
SN - 0169-023X
VL - 116
SP - 42
EP - 60
JO - Data and Knowledge Engineering
JF - Data and Knowledge Engineering
ER -