MMSVC: An Efficient Unsupervised Learning Approach for Large-Scale Datasets

• Published in: Life System Modeling and Intelligent Computing pp. 1 - 9
• Main Authors: Gu, Hong, Zhao, Guangzhou, Zhang, Jianliang
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg
• Series: Lecture Notes in Computer Science
• Subjects: large-scale clustering; one-class support vector machine; support vector clustering
• ISBN: 3642156142; 9783642156144
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-642-15615-1_1

This paper presents a multi-scale, hierarchical framework to extend the scalability of support vector clustering (SVC). Based on the multi-sphere support vector clustering, the clustering algorithm called multi-scale multi-sphere support vector clustering (MMSVC) in this framework works in a coarse-to-fine and top-to-down manner. Given one parent cluster, the next learning scale is generated by a secant-like numerical algorithm. A local quantity called spherical support vector density (sSVD) is proposed as a cluster validity measure which describes the compactness of the cluster. It is used as a terminate term in our framework. When dealing with large-scale dataset, our method benefits from the online learning, easy parameters tuning and the learning efficiency. 1.5 million tiny images were used to evaluate the method. Experimental results demonstrate that the method greatly improves the scalability and learning efficiency of support vector clustering.