The increasing usage of Geographical Information Systems (GIS) for various problems makes the volume of spatio-temporal data. There is an urgent need for effective and efficient methods to extract unknown and unexpected information from spatiotemporal data sets of exceptionally large size, high dimensionality and complexity. In that forming the rules, generating the characteristics and discovering the trends accurately are very difficult. To address these challenges, the authors propose a new algorithm to get better results with different types of spatio-temporal datasets in different number of iterations. In this paper, clustered and outliers removed spatio-temporal data were used. Rule generation is to predict the value of one attribute based on another attribute over time and it generalizes the characteristics of the generated rules.