Computing and Informatics

In this paper we compare usefulness of statistical techniques of dimensionality reduction for improving clustering of documents in Polish. We start with partitional and agglomerative algorithms applied to Vector Space Model. Then we investigate two transformations: Latent Semantic Analysis and Probabilistic Latent Semantic Analysis. The obtained results showed advantage of Latent Semantic Analysis technique over probabilistic model. We also analyse time and memory consumption aspects of these transformations and present runtime details for IBM BladeCenter HS21 machine.