Development of decision rules for site-specific N fertilization by the application of data mining techniques
Georg Weigert, Péter Wagner · 2003
An integrated system using site-specific data for the development of decision rules for variable rate N Fertilization is proposed. In the paper, each step of this process which consists of data gathering, data storing, data processing and data analysis is presented. The emphasis is placed on performing data analysis by Data Mining Methods. Initial results are promising. A neural network trained with single year data showed high accuracy in predicting yield. It also showed that the major part of decision rules could be soil electrical conductivity, vegetation index and draft force.