Semi-Supervised Cluster Ensemble for Web Video Categorization

• Affiliation of Author(s):西南交通大学
• Journal:11th International Conference on Multiple Classifier Systems
• Key Words:Clustering, Cluster Ensemble, Pairwise Constraints, Video Categorization.
• Abstract:Recently, web video categorization has been an ever interesting research with the popularity of web videos. Clustering ensemble has become a good alternative for categorization. Semi-supervised clustering ensemble has shown a better performance since it may incorporate the known prior knowledge, e.g., pairwise constraints. In this paper, we propose a Semi-supervised Cluster-based Similarity Partitioning Algorithm (SS-CSPA) to categorize the videos containing textual data provided by their up-loaders. The feature of this algorithm is the introduction of an unsupervised learning, consensus between clustering and additional support of pairwise constraints to formulate semi-supervised clustering ensemble paradigm. Experimental results on the real-world web videos show that the proposed algorithm outperforms existing methods for categorization of web videos.
• Co-author:Tianrui Li, Yan Yang, Hongjun Wang
• First Author:Amjad Mahmood
• Indexed by:Academic papers
• Document Code:20140517252230
• Discipline:Engineering
• First-Level Discipline:Computer Science and Technology
• Issue:LNCS 7872
• Page Number:pp. 190–200, 2013.
• Translation or Not:no