Natural Nearest Neighbor for Isomap Algorithm without Free-Parameter

Abstract:

Article Preview

Isomap is a classic and efficient manifold learning algorithm, which aims at finding the intrinsic structure hidden in high dimensional data. Only deficiency appeared in this algorithm is that it requires user to input a free parameter k which is closely related to the success of unfolding the true intrinsic structure and the algorithm’s topological stability. Here, we propose a novel and simple k-nn based concept: natural nearest neighbor (3N), which is independent of parameter k, so as to addressing the longstanding problem of how to automatically choosing the only free parameter k in manifold learning algorithms so far, and implementing completely unsupervised learning algorithm 3N-Isomap for nonlinear dimensionality reduction without the use of any priori information about the intrinsic structure. Experiment results show that 3N-Isomap is a more practical and simple algorithm than Isomap.

Info:

Periodical:

Advanced Materials Research (Volumes 219-220)

Edited by:

Helen Zhang, Gang Shen and David Jin

Pages:

994-998

DOI:

10.4028/www.scientific.net/AMR.219-220.994

Citation:

X. L. Zou et al., "Natural Nearest Neighbor for Isomap Algorithm without Free-Parameter", Advanced Materials Research, Vols. 219-220, pp. 994-998, 2011

Online since:

March 2011

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.