Implementasi Particle Swarm Optimization pada Sistem Rekomendasi Tanaman Hortikultura Berbasis Naïve Bayes

Abstract

Indonesia is an agrarian country with an environment that strongly supports agricultural processes. Because of this, crops are one of Indonesia’s main commodities and choosing a right crop to cultivate becomes a crucial process. Nowadays, lots of recommendation system has been made to help with decision making. A crop recommendation system is one of them and prove to be helpful in helping farmers decide what crop to plant based on the condition of the environment. An idea of implementing particle swarm optimization on a crop recommendation system occurred. Particle swarm optimization can be implemented to choose the presumably best parameter for a Gaussian Naïve Bayes model. The result of implementing particle swarming optimization to find the best smoothing parameter is the accuracy of the model reaching 99.5%. However, this is not different to the result of the model without particle swarm optimization which reach the accuracy of around 99.5% too.