The fuzzy system technology in geo-spatial data mining

Although data mining is a relatively young technique, it has been used in a wide range of problem domains during the past few decades. In this paper, the authors present a new model that applies the data mining technique to forecasting the demand for cultivated land. The new model is called the fuzzy Markov chain model with weights. It applies data mining techniques to extract useful information from the enormous quantities of historical data and then applies the fuzzy sequential cluster method to set up the dissimilitude fuzzy clustering sections. The new model regards the standardized self-correlative coefficients as weights based on the special characteristics of correlation among the historical stochastic variables. The transition probabilities matrix of the new model is obtained by using fuzzy logic theory and statistical analysis. The experimental results show that the ameliorative model, combined with the technique of data mining, is more scientific and practical than traditional predictive models.