A data mining application on air temperature database

Abstract

In this study, a data mining application based on DBSCAN (Density Based Spatial Clustering of Applications with Noise) was carried out on air temperature database which contains daily temperature data from country wide meteorology stations in Turkey. At the end of data mining process, we obtained clusters that have similar temperature trends. These clusters have been used to categorize Turkey into regions according to climatic characteristics. Statistical methods are widely used in meteorology; however they need extreme computing power. Data mining methods provide more performance and reliability than statistical methods.