Soil Classification Using GATREE

Abstract

This paper details the application of a geneticprogramming framework for classification of decisiontree of Soil data to classify soil texture. Thedatabase contains measurements of soil profile data. Wehave applied GATree for generating classificationdecision tree. GATree is a decision tree builder thatis based on Genetic Algorithms(GAs). The idea behind itis rather simple but powerful. Instead of usingstatistic metrics that are biased towards specifictrees we use a more flexible, global metric of treequality that try to optimise accuracy and size. GATreeoffers some unique features not to be found in anyother tree inducers while at the same time it canproduce better results for many difficult problems.Experimental results are presented which illustrate theperformance of generating best decision tree forclassifying soil texture for soil data set.