Fuzzy c-medoids method based on JS-divergence for uncertain data clustering
- Resource Type
- Conference
- Authors
- Wang, Yingxu; Dong, Jiwen; Zhou, Jin; Wang, Dong; Wang, Lin; Han, Shiyuan; Chen, Yuehui
- Source
- 2017 4th International Conference on Information, Cybernetics and Computational Social Systems (ICCSS) Information, Cybernetics and Computational Social Systems (ICCSS), 2017 4th International Conference on. :312-315 Jul, 2017
- Subject
- Communication, Networking and Broadcast Technologies
Engineered Materials, Dielectrics and Plasmas
General Topics for Engineers
Robotics and Control Systems
Signal Processing and Analysis
Clustering algorithms
Random variables
Gaussian distribution
Measurement uncertainty
Algorithm design and analysis
Linear programming
Cybernetics
uncertain data clustering
JS-divergence
Fuzzy c-medoids method
- Language
Uncertain data clustering is one significant research in data mining. Many similarity measurements of uncertain objects are proposed. Traditional clustering methods can be extended with these new similarity measurements. In this paper, we propose a new fuzzy c-medoids method for uncertain data clustering, named UFC-medoids. The JS-divergence is used as the similarity measurement between uncertain objects in this algorithm. In the experiments on synthetic datasets, the presented algorithm has shown a good performance.