DAILY STREAMFLOW PREDICTION ON TIME SERIES FORECASTING

Time series forecasting is a process that used present or past data to develop models for future prediction or trends. Stream flow prediction is considered as a challenging research activity because of its irregularity and unpredictable behavior. Researches have put their efforts and strategies in upgrading and improving the accuracy of streamflow analysis prediction. In this paper, time series forecasting using WEKA is used, analyzed and compared based on the following three algorithms, which are SMO Regression, Linear Regression and Multilayer Perception. The result shows that the SMO Regression algorithm provides better ability to predict more accurately compared to other algorithms.