'This thesis document deals with the context of data mining. We explain various clustering techniques such as partition-based techniques and probabilistic approaches.

We have implemented the k-means clustering algorithm, the trimmed k-means variation of it and the Naïve Bayes with EM learning for clustering. We have compared the results of these algorithms applied to the given database.

Furthermore, we have implemented a tool where these clustering techniques can be applied, and the results of those techniques can be shown on a map of the geographical area where the data come from. '