Incremental conic functions algorithm for large scale classification problems
DIGITAL SIGNAL PROCESSING, vol.77, pp.187-194, 2018 (SCI-Expanded, Scopus)
- Publication Type: Article / Article
- Volume: 77
- Publication Date: 2018
- Doi Number: 10.1016/j.dsp.2017.11.010
- Journal Name: DIGITAL SIGNAL PROCESSING
- Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
- Page Numbers: pp.187-194
- Keywords: Polyhedral conic functions, Mathematical programming, Classification, Machine learning, SEPARATION, SEPARABILITY
- Anadolu University Affiliated: Yes
Abstract
In order to cope with classification problems involving large datasets, we propose a new mathematical programming algorithm by extending the clustering based polyhedral conic functions approach. Despite the high classification efficiency of polyhedral conic functions, the realization previously required a nested implementation of k-means and conic function generation, which has a computational load related to the number of data points. In the proposed algorithm, an efficient data reduction method is employed to the k-means phase prior to the conic function generation step. The new method not only improves the computational efficiency of the successful conic function classifier, but also helps avoiding model over-fitting by giving fewer (but more representative) conic functions. (C) 2017 Elsevier Inc. All rights reserved.