Clustering analysis in search of wind impacts on evaporation

Abstract

Clustering deals with finding a structure in a collection of uncategorized data and can be examined the most important unsupervised learning problem and the other problems as kind of this. The aim of this study is to cluster the monthly evaporation losses with the monthly winds speed and wind blow number of Egirdir Lake, one of the most important fresh water storage of Turkey. For this aim, wind speed and evaporation data also wind blow number depend on hourly and daily mean records measured in Egirdir Lake Catchment, are used. In the clustering analysis of the data, as a non-parametric approach hierarchical clustering algorithm was successfully applied at different similarity stages. © 2008, Penkala Bt.