Pthread Parallel K-means

K-means is a popular non-hierarchical method for clustering large datasets. The time requirements increase linearly with the size of the data set which make it particulary suited for extremely large datasets such as those found in digital libraries. The method was developed by MacQueen [4] in 1967. In our project we take a uniprocessor k-means algorithm and implement a parallel k-means algorithm using pthreads. The algorithm we use is from the Normalized Cuts (Ncuts) [6] codebase of the Vision Group at UC Berkeley. The rest of this paper is organized as follows: We start by providing information about Ncuts and kmeans. In Section 3.2, we describe our software testbed, and present our design and implementation of the parallel k-means algorithm. Then in Section 3.3 we show our performance measurements and results. We conclude and discuss our plans for future work in Section 4.