#Local-Kernel-Alignment-Based-Multi-view-Clustering-Using-ELM The contributions of this study can be summarized as follows: (a) This study proposes a general multi-view clustering approach which com- bines multiple views generated by ELM random feature mapping on original single-view dataset. Each view corresponds to a different value of hidden-layer nodes.

(b) This study exploits the properties of the local kernel alignment of con- structed views and proposes an ELM-based multiple kernel clustering algorithm with local kernel alignment maximization.

(c) This study presents state-of-the-art clustering results on 10 datasets, including two datasets used in practical face recognition tasks.