Sequential Regression Models with Pairwise Constraints Using Noise Clusters
Hengjin Tang, Sadaaki Miyamoto · Journal of Advanced Computational Intelligence and Intelligent Informatics · 2012
Switching regression models are useful in a variety of real applications. Semi-supervised clustering with pairwise constraints is also well-known to be important and many researchers recently study this subject. In spite of their usefulness, there is one drawback: the results have a strong dependency on the predefined number of clusters. To avoid this drawback, we use a method of sequentially extracting one cluster at a time using noise-detecting method, and propose constrained switching regressionmodels which enables an automatic determination of clusters. We show the effectiveness of the proposed method by using numerical examples.