Abstract:
Weather prediction is an essential activity in today's world economy with its detrimental effects on various fields like Agriculture, Utility companies, Marine etc. Many methods have been presented based on fuzzy time series to make predictions in areas such as stock price, university enrolments, weather, etc. When using fuzzy time series for forecasting, the length of intervals in the universe of discourse is important due to the fact that it can affect the forecasting accuracy rate. This paper proposes a better approach to forecasting temperature by applying automatic clustering algorithm to partition the universe of discourse. Improvement in results is observed as compared to existing techniques that involve partitioning the universe of discourse in static intervals. The proposed method is tested on temperature prediction and improvements in results are compared to some of already existing techniques.