An Association Rule Mining Approach for Satellite Cloud Images and Rainfall.

This paper aims at discovering useful knowledge from a large collection of satellite cloud images and rainfall data using image mining. The paper illustrates how important the data conversion is in building an accurate data mining architecture. Most of data about image features and rainfall data are values or vectors, which are not fit for mining directly. We present two approaches to implement the conversion of data: a clustering algorithm and a fuzzy clustering method (FCM). The clustering algorithm is used to map the numerical value to categorical value. The FCM implements the conversion of feature vector. Finally, the association rules are determined using the Apriori algorithm. The experiment results show that the acquired association rules are consistent with the fact and the results are satisfying. Keywords: Association rules mining, satellite cloud image, clustering partition. [ABSTRACT FROM AUTHOR]