We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
A Novel Spectral Clustering based on Local Distribution.
- Authors
Roy, Parthajit; Mandal, J. K.
- Abstract
This paper proposed variation of spectral clustering model based on a novel affinity metric that considers the distribution of the neighboring points to learn the underlaying structures in the data set. Proposed affinity metric is calculated using Mahalanobis distance that exploits the concept of outlier detection for identifying the neighborhoods of the data points. Random Walk Laplacian of the representative graph and its spectra has been considered for the clustering purpose and the first k number of eigenvectors have been considered in the second phase of clustering. The model has been tested with benchmark data and the quality of the output of the proposed model has been tested in various cluster validity indexes.
- Publication
International Journal of Electrical & Computer Engineering (2088-8708), 2015, Vol 5, Issue 2, p361
- ISSN
2088-8708
- Publication type
Academic Journal