The project represents the machine learning-based predictive analysis to predict the rainfall for upcoming day based on the given data. The proposed system will help to predict the future trend of the weather considering the historical data. Through this model, the probability of the rainfall can be analyzed so that all the people can make their plans without any confusion.
I have Proposed an experimental approach to develop a Rainfall prediction classifier to predict future weather condition based on the other features. The best features will be selected on the basis of their linear relationships. Furthermore, the following baseline ML classifiers and deep Sequential model will be trained on the weather data.
- Logistic Regression
- K-Nearest Neighbors
- Decision Tree
- AdaBoost Tree
- Random Forest
- Sequential Model
The models will be trained using the weather data and predict the RainTomorrow attribute from the unknown weather data to identify the weather condition. Furthermore, the best-performing model will be improved to find the optimized configuration of the parameters for achieving better classification accuracy on the validation data. Finally, the RainTomorrow attribute will be predicted for the unknown weather data using the optimized classifier.