Abstract
This study is aimed to evaluate the applicability of data mining technique by using decision tree in land evaluation. It can be used to determine the land characteristic factors affecting the agricultural land-use potential and quantify the relationship between land characteristic factors and plant productivity in order to improve land evaluation methods that support the foundation of land use planning. Regression decision tree model in this study includes two kinds of variables. The target variable is the productivity (t/ha) and the predictor variables consist of soil types, soil depth, slope, irrigation and texture. The analytical result of survey data shows several factor combinations according to plant average productivity. Based on productivity can evaluate the adaptation level for each correlative factor combination. This study is applied for rubber trees and conducted in Phu Giao district, Binh Duong province. The study shows that the interpretation level of the predictive variables is 96.49%. The area of highly suitable (S1) is 474.67 hectares, suitable (S2) is 53,597.70 hectares. This result is different from the Analytic Hierarchy Process (AHP) method.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
