Article Preview
TopMany researchers from different fields worked on temperature forecasting using different techniques, Al-Matarneh et.al, applied two different models for temperature forecasting, Feed Forward Neural Networks with back propagation algorithm and Fuzzy Logic model. Different evaluation criteria were used, the Variance Accounted For (VAF), and Mean Absolute Error (MAE). The results obtained were good and showed that the proposed models can act with more accuracy.
Radhika and Shashi, used the Support Vector Machines (SVMs) to predict the maximum weather temperature at a particular location based on the daily time series observations.
Hayati et.al, used the Artificial Neural Network with Multilayer perceptron (MLP) to predict the temperature for ten years dataset (1996 to 2006). The dataset was divided into two sets one for training and other for testing. The performance of the MLP network was very good with minimum errors.
Patel and Christian, developed a temperature forecasting model for Inland Cities in India. The dataset of one year of daily temperature observations was considered. Relative humidity and mean sea level pressure were measured as inputs variables. The obtained results showed a great decrease in the Root mean square error.