| dc.description.abstract | Forest has many benefits for human life. Nowadays forest areas in Indonesia have decreased because of illegal logging, forest fires and forest conversion. Wildfires have resulted economic losses, health problems and pollution. Forest fires can be indicated through hotspot occurrences. In this work, data modeling was conducted to predict hotspot occurrences using Autoregressive Integrated Moving Average (ARIMA). ARIMA is one of prediction methods that can be used for modeling timeseries data such as hotspots that are daily recorded by satellite sensor. Modeling was performed on monthly hotspots occurrences data for the period of 2001 to 2012 in Riau Province. The experimental results showed the ARIMA(2,0,0) model was the best model to predict the number of monthly hotspot occurrences with a Mean Absolute Percentage Error (MAPE) of 40.974. | en |
