[Back]


Talks and Poster Presentations (with Proceedings-Entry):

T. Dittrich, P. Berger, G. Matz:
"Semi-Supervised Spectral Clustering using the Signed Laplacian";
Poster: Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, USA; 10-28-2018 - 10-31-2018; in: "51st Asilomar Conference on Signals, Systems, and Computers", (2018), 5 pages.



English abstract:
Data clustering is an important step in numerous real-world problems. The goal is to separate the data into disjoint subgroups (clusters) according to some similarity metric. We consider spectral clustering (SC), where a graph captures the relation between the individual data points and the clusters are obtained from the spectrum of the associated graph Laplacian. We propose a semi-supervised SC scheme that exploits partial knowledge of the true cluster labels. These labels are used to create a modified graph with attractive intra-cluster edges (positive weights) and repulsive inter-cluster edges (negative weights). We then perform spectral clustering using the signed Laplacian matrix of the resulting signed graph. Numerical experiments illustrate the performance improvements achievable with our method.

Keywords:
graph signal processing, semi-supervised clustering, spectral clustering


Electronic version of the publication:
https://publik.tuwien.ac.at/files/publik_274228.pdf


Created from the Publication Database of the Vienna University of Technology.