Addressing the importance of uncertainty in climatic predictions based on statistical methods

As there are climatic changes, there are many risks involved in predicting these changes and make decisions on some infrastructure due to extreme weather conditions and shifting weather patterns. Existing research indicates that small increase in the climate and weather extremes which have the potential to bring large calamities to the existing infrastructures. These changes may be due to the presence of uncertainties in values. Many researches were on predicting and it has been a challenging field, so many algorithm deals with it. Such uncertainties are predicted and evaluated using many techniques. In this paper we have predicted uncertainty with some statistical methods like mean and standard deviation on the weather dataset. The error rate, accuracy level makes the system efficiency, which is done in this paper using the statistical measures and removes the uncertainty value in the dataset. The Support Vector Machine (SVM) classifier improves the classification accuracy thus in turn improves the prediction rate accuracy.