Optimization of Land Suitability for Food Crops Using Neural Network and Swarm Optimization Algorithm
Rita Layona, Suharjito Suharjito, Yovita Tunardi, Dian Felita Tanoto · International Review on Computers and Software (IRECOS) · 2016
Land quality and suitability are factors that affect food crop productivity. Inappropriate or unproductive land will impact to productivity decrease, time-consuming, and profit lost. Determining and evaluating the optimal land suitability can be done through prediction of each production area that is obtained by analyzing land fitness data. This study proposes the implementation of Neural Network and Swarm Optimization Algorithm (Particle Swarm Optimization & Cat Swarm Optimization) to obtain prediction of production. To evaluate the proposed method, this study performed comparative evaluation based on the Mean Square Error (MSE) and accuracy. Based on the experimental results, Cat Swarm Optimization produces minimum error equal of 0.00439 for training phase and 0.10453 for testing phase. In training phase, Cat Swarm Optimization produces higher accuracy (93%) than Particle Swarm Optimization (67%).