The Mining of the Real Valued Datasets using Clustering Algorithms: Review Paper

Cluster analysis has been widely used in several disciplines, such as statistics, software engineering, biology, psychology and other social sciences, in order to identify natural groups in large amounts of data. These data sets are constantly becoming larger, and their dimensionality prevents easy analysis and validation of the results. The subspace pattern mining has been tailored to microarray data clustering to find biclusters and triclusters. We focus on deterministic clustering algorithm: Triclusters, which can mine arbitrarily positioned and overlapping biclusters/triclusters. Depending on different parameter values, they can mine different types of clusters, including those with constant or similar row/column values, as well as scaling and shifting expression patterns. We also give a useful set of metrics to evaluate the clustering quality, and show their effectiveness on real data