A Novel Decision Tree Approach for the Prediction of Precipitation Using Entropy in SLIQ

The rising tendency of population of every nation is one of the severe stumbling blocks to arrest its economic growth, particularly in the third world countries like India, not able to address even the basic needs of its people. It is high time to have introspection for the deficiencies and find a remedy. The major basic need is food, a product of agriculture. Agriculture mainly depends on rainfall. Prediction of precipitation is a complex phenomenon. Till now many of the researchers have tried their best for predicting the precipitation but in vain, since the prediction is quite complex with neural networks, back propagation, fuzzy logic etc. Hence, we found that data mining is an emerging, efficient, easily implementable tool, which predicts the useful patterns for the prediction of rainfall in a very short time. Supervised Learning In Quest, an efficient data mining decision tree algorithm is applied in the prediction of precipitation. The present research illustrates the Supervised Learning In Quest decision tree algorithm using entropy, which estimates the prediction of precipitation with an average accuracy of 74.92% and the knowledge extraction is purely based on historical data.