Title :
An improved cluster labeling method for support vector clustering
Author :
Lee, Jaewook ; Lee, Daewon
Author_Institution :
Dept. of Ind. Eng., Pohang Inst. of Sci. & Technol., South Korea
fDate :
3/1/2005 12:00:00 AM
Abstract :
The support vector clustering (SVC) algorithm is a recently emerged unsupervised learning method inspired by support vector machines. One key step involved in the SVC algorithm is the cluster assignment of each data point. A new cluster labeling method for SVC is developed based on some invariant topological properties of a trained kernel radius function. Benchmark results show that the proposed method outperforms previously reported labeling techniques.
Keywords :
pattern clustering; statistical analysis; support vector machines; topology; unsupervised learning; cluster assignment; cluster labeling method; kernel radius function; support vector clustering algorithm; support vector machines; topological properties; unsupervised learning method; Character generation; Clustering algorithms; Computer simulation; Kernel; Labeling; Robustness; Shape; Static VAr compensators; Support vector machines; Unsupervised learning; Index Terms- Clustering; support vector machines.; unsupervised learning method; Algorithms; Artificial Intelligence; Cluster Analysis; Information Storage and Retrieval; Numerical Analysis, Computer-Assisted; Pattern Recognition, Automated; Reproducibility of Results; Sensitivity and Specificity;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2005.47
