Instance selection based on supervised clustering

Instance selection is one of important steps in pattern classification. Recently, instance selection is a hot research topic in pattern recognition, data mining, machine learning, and draws many researchers´ attention. By instance selection, we can eliminate the redundant instances in the datasets, and select more important and fewer samples as training set to train a classifier with good generalization performance. In this paper, we present an instance selection method based on supervised clustering, the main idea is to select instances belonging to inner boundary and outer boundary of clusters. The experimental results show that our proposed method is effective and efficient.