A cloud computing framework for analysis of agricultural big data based on Dempster–Shafer theory

This paper aims to extract optimal location for cultivating orange trees. In order to reach this goal, a combination of Dempster-Shafer theory (DST) and cloud computing is proposed. The DST method is applied to make weights for input parameters, and cloud computing is used for creating a cost-effective integrated solution on collected information of different geographic regions. To do this, eight parameters including minimum and maximum temperatures, aspect, elevation, growing degree days, rainfall, relative humidity, solar radiation and slope are incorporated to determine the most optimal region for orange cultivation. Moreover, interpolation maps for each parameter are determined with using the inverse distance weighting model in a geographic information system software. The DST model as a novel method for the determination of land suitability is eventually applied in the MATLAB software environment to complete the performance evaluation. Three confidence levels are set as 99.5%, 99% and 95% such that the final results for each confidence level are compared accordingly. It will be shown that the proposed method is successful in predicting suitable locations for the cultivation of oranges by generating different maps at various degrees of confidence.