MSSS

Sampling with Minimum Sum of Squared Similarities for Nystrom-Based Large Scale Spectral Clustering Publication

Project Description

The Nystrom sampling provides an efficient approach for large scale clustering problems, by generating a low-rank matrix approximation. However, existing sampling methods are limited by their accuracies and computing times. Here we propose a scalable Nystrom-based clustering algorithm with a new sampling procedure, called: Minimum Sum of Squared Similarities (MSSS).

Publication

• Bouneffouf D., Birol I.: Sampling with Minimum Sum of Squared Similarities for Nystrom-Based Large Scale Spectral Clustering.  Proceedings of the Twenty-Fourth international joint conference on Artificial Intelligence (IJCAI) , 2015-July.