Price Prediction of Chili in Bandung Regency Using Support Vector Machine (SVM) Optimized with an Adaptive Neuro-Fuzzy Inference System (ANFIS)

Abstract

The price fluctuation of chili is one of the economic problems faced by every chili-producing country in the world, including in Indonesia. Chili is a vegetable that is consumed almost every day by the people of Indonesia. In Bandung district area, chili has been experiencing price fluctuations in the last four years, according to data obtained from the Bandung Regency Area Trade and Industry Service. Many factors cause the price of chili to fluctuate—one of them being the weather. This is because chili is a plant that is easily damaged if exposed to too much water. This research predicts chili prices in Bandung Regency using the Support Vector Machine (SVM) algorithm, which is optimized by an Adaptive Neuro-Fuzzy Inference System (ANFIS) and based on weather factors. The average accuracy of training and testing data was 94.07%; the training and testing data using the SVM algorithm produced 89.90% average accuracy and the average accuracy of training and testing data using the ANFIS algorithm was 92.68%.