Prediksi Curah Hujan Menggunakan Metode Adaptive-Expectation Based Multi-Attribute Fuzzy Time Series

Yulia Trianandi (Fakultas Ilmu Komputer, Universitas Brawijaya)
Wayan Firdaus Mahmudy (Fakultas Ilmu Komputer, Universitas Brawijaya)
Candra Dewi (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 2 No 12 (2018): Desember 2018

Prediction of rainfall is needed in increasing the production of food crops in a region, one of them is Tengger area. Errors in predicting rainfall can cause errors when determining the planting period, and the right type of plant. In order to produce rainfall predictions with a slight error rate, this study uses the Adaptive-Expectation Based Multi-Attribute Fuzzy Time Series method. The method has been proven to predict the closing price on the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAEIX) with fewer error rates than Chen Fuzzy Time Series methods. This study will produce rainfall predictions in the four districts of Tengger area, namely Puspo, Sumber, Tosari, and Tutur. From the results of testing 36 data in 2014, obtained the best Mean Square Error (MSE) value of 28.0470 in Tosari district.