Expert System for Land Suitability Evaluation using Data mining ‘ s Classification Techniques : a Comparative Study

Abstract

Data mining involves the extraction of implicit, “interesting” information from a database. Classification is an important Data mining’s “machine learning” technique which is used to predict data instances from dataset. It involves the order wise analysis of large amount of information sets. Data mining applications are used in various areas such as health care, insurance, medicines, Agriculture, banking and soil management. In soil region the Data mining mainly used to classify the soil and predicting the land suitability for the crop and fertilizer recommendation. The purpose of this study is to predict the land suitability for the crop using classification algorithms namely Naive Bayes and J48. This work focused on find out the best classification algorithm based on accuracy measure, performance measure, error rate and execution time using the soil dataset. From the experimental result using WEKA tool it is observed that the performance of the J48 is better than the Naive Bayes algorithm.