Added ML project on PGA prediction of Earthquake data

[subhmm]

Iran is one of the most seismic countries of the world. It is situated over the Himalayan-Alpide seismic belt and is one of those countries, which have lost many human lives and a lot of money due to the occurrence of earthquakes. Here a model is built using Machine Learning to predict PGA in this region.

The dataset was split into input part and the PGA value which was supposed to be predicted. The input part was standardised using ​StandardScalar. ● Now, the entire dataset was split into ​test set ​and​ training set. ● We trained the model using ​Logistic Regression, K-Nearest Neighbours(KNN) ​and Random Forest Regression ​using the training set. ● The trained models were tested with the test set to ​predict the efficiency​ of the model. ● Finally, two ​Ensemble models ​ were created combining the previous models, one is the Averaging​ technique and the other is ​Blending​ technique. The efficiency of the models were predicted by calculating the Mean Average Error​ MAE ​ , Mean Squared Error- ​ MSE ​ and Root Mean Squared Error- R ​ MSE. Lower the value, the higher is the efficiency.