The knowledge of land capability classification (LCC) is a prerequisite and important for planning of agriculture development programs. The capability of land depends upon the characteristics of the land such as texture, soil depth, topography, salinity, drainage, calcium carbonate and so on. Considering these input parameters, (LCC) decision tree was utilized usingIterating Dichotomizer 3 algorithm (ID3) for evaluating the land capability classification of the main physiographic units of some soils of the Eastern Desert Part of Sohag Governorate. A 10 fold cross validation provided an accuracy of 91.67%.It is concluded that the model is useful for suggesting that explicit rules could be formulated with better accuracy for classifying complex soil-site data acquired over diversified land types.