The Homogeneous Ensemble Methods for MLknn Algorithm
Khalida Douibi, Nesma Settouti, Mohammed El Amine Chikh · 2017
The multi-label classification is one of the current research orientation since it solves many real problems where each object can have several semantics. One of the categories dedicated for learning from such data is the adaptation methods. In this paper, we propose to improve the performance of the Multi-label K Nearest Neighbors (MLknn) using the ensemble methods (Bagging and Boosting), it adapts the K Nearest Neighbors algorithm to Multi-label data. The experimental results on five small to larger multi-label datasets from different domains, shows the effectiveness of the ensemble methods to improve the original algorithm.