Price Prediction Model

The House Price Prediction project used the data near the Californian region. The dataset was visualised using various scatterplots and histograms, it was then preprocessed - imputation (SimpleImputer), encoding (ordinal, one-hot, label encoding), feature scaling (StandardScalar, MinMaxScalar), use of radial basis function (RBF) to compute similarity, transformation pipelines, trained using linear regression, decision tree regressor, fine-tuning via Grid Search and Randomized search and finally evaluated on test set. The median income was the most useful predictor.