Guo, Y

Gao, J

Kwan, PH

2006

Visualization of non-vectorial objects is not easy in practicedue to their lack of convenient vectorial representation.Representative approaches are Kernel PCA and KernelLaplacian Eigenmaps introduced recently in our research.Extending our earlier work, we propose in this papera new algorithm called Twin Kernel Embedding (TKE)that preserves the similarity structure of input data in the latentspace. Experimental evaluation on MNIST handwrittendigit database verifies that TKE outperforms related methods.

Proceedings of The 2006 International Workshop on Integrating AI and Data Mining (AIDM'06), p. 11-17

0769527302

https://hdl.handle.net/1959.11/1021

Institute of Electrical and Electronics Engineers (IEEE)

Visualization of Non-vectorial Data Using Twin Kernel Embedding

Conference Publication

Publication