IMPROVED FUZZY K-MEANS CLUSTER ALGORITHM TO ANALYSE WEATHER DATA IN COIMBATORE REGION

Abstract

Weather analysis has been playing its vital role in meteorology and become one of the most challengeable
problems both scientifically and technologically all over the world from the recent years. This research paper carries
historical weather data collected locally at South Indian textile city, Coimbatore that was analyzed for useful knowledge by
applying data mining methods. Data includes last five years’ period [2012-2017]. It had been tried to extract useful practical
knowledge of weather data on monthly based historical analysis. This Proposed research work was done using data mining
tool called Weka by examining changing patterns of weather parameters which includes maximum temperature, minimum
temperature, wind speed and rainfall. After preprocessing of data and outlier analysis, improved fuzzy K-means clustering
algorithm and Decision Tree techniques were applied. Two clusters were generated by using improved K-means Clustering
algorithm with lowest and highest of mean parameters. The result obtained with smallest error (27%) was selected on test
data set. While for the number of rules generated of the given tree was selected with minimum error of 20%. The results
showed that for the given adequate set data, these techniques can be used for weather investigation and climate change
studies.