Pre-Processing and Normalization of the Historical Weather Data Collected from Secondary Data Source for Rainfall Prediction

In the twenty first century, data analysis has become the talk of the town. Almost every company or organization depends on data analysis for taking future decision. The most important step in data analysis after data collection is the preprocessing of the collected data. The main aim of data analysis is to find meaningful pattern by processing large amount of data. In data preprocessing, the inconsistency of collected data has been removed. After storing data for a relatively longer period, it becomes noisy and inconsistent. While measuring various parameter due to error in the instrument or human error, the value become incorrect or invalid. It is necessary to remove the invalid data otherwise it will deflect the results and produce error in the prediction. In this work preprocessing of the weather data has been analyzed for rainfall prediction using data mining.