The Fuzzy Inference System with rule bases generated by using the Fuzzy C-Means to predict regional minimum wage in Indonesia
International Journal of Operations and Quantitative Management
International Forum of Management Scholars
School of Business and Law / Centre for Innovative Practice
The aims of the research are to build the Fuzzy Inference System (FIS) Takagi-Sugeno of zero order and to investigate the influence number of linguistic value on FIS performance evaluated by using MAPE and R2 indicator. The fuzzy rule bases are generated by using the Fuzzy C-Means (FCM) clustering. The regional minimum wage in 2015 and 2016 dataset of East Java province of Indonesia are consecutively used as the training and the testing data. The number of clusters (n) that become FCM input including n = 3, 5, 7, and 9. The fuzzification process uses the Gaussian membership function with the mean parameter is the cluster center of the FCM output. Based on the values of MAPE and R2 on the training and the testing data is obtained the system with the number of cluster of 9 which is the best performance. Furthermore, the increasing number of cluster is followed by the decreasing MAPE and the increasing R2 value in both training and testing data.