Modeling farmers intensication decision requires a model that considers the dependencies between the perceived influences and their choices of intensication pathways, accounting uncertainties at the same time. A combination of data-driven Bayesian Belief Network (BBN) and Regression Tree is proposed in this paper. Data from 304 rural households in Kilombero Valley Floodplain in Tanzania is used to learn the structure and parameter of the model. The resulting BBN is able to drive the probabilities of intensication choices conditional on key market, biophysical and socio-economic characteristics of farm households. Acknowledgement : This research was conducted under the GlobE Wetlands project, which is funded by the German Federal Ministry of Education and Research (FKZ: 031A250 A-H), with additional funding provided by the German Federal Ministry for Economic Cooperation and Development.