Title :
Cluster-dependent feature selection by multiple kernel self-organizing map
Author :
Kuan-Chieh Huang ; Yen-Yu Lin ; Jie-Zhi Cheng
Author_Institution :
Res. Center for Inf. Technol. Innovation, Taipei, Taiwan
Abstract :
Motivated by the fact that data of each cluster are often well captured by distinct features, we propose a clustering approach called multiple kernel self-organizing map (MK-SOM) that integrates multiple kernel learning into the learning procedure of SOM, and carries out cluster-dependent feature selection simultaneously. MK-SOM is developed to reveal the intrinsic relation between features and clusters, and is derived with an efficient optimization procedure. The proposed approach is evaluated on two benchmark datasets, UCI and Caltech-101. The promising experimental results demonstrate its effectiveness.
Keywords :
data analysis; learning (artificial intelligence); optimisation; pattern clustering; self-organising feature maps; Caltech-101 datasets; MK-SOM; UCI datasets; benchmark datasets; cluster-dependent feature selection; data analysis problems; intrinsic relation; multiple kernel learning; multiple kernel self-organizing map; optimization procedure; Clustering algorithms; Clustering methods; Kernel; Machine learning; Optimization; Training; Vectors;
Conference_Titel :
Pattern Recognition (ICPR), 2012 21st International Conference on
Conference_Location :
Tsukuba
Print_ISBN :
978-1-4673-2216-4