/Multi_linear_regression-Comprehensive-way-of-implementation

Prediction of Boston houses price by Multilinear regression using different data science and machine learning aspects.

Primary LanguageJupyter Notebook

Multi_linear_regression-Comprehensive-way-of-implementation

Steps implemented in notebook are given below:

  1. Import dependencies

  2. Import Dataset

  3. Dataframe transformation and functions on dataset

  4. Data distribution of data frames (Dataset)

  5. Model Fitting

    Standardization (Optional)

    Normalization (Optional)

  6. Prediction

  7. Comparison with actual values

  8. Model Evaluation Metrics:

    Mean Square Error (MSE)

    R2-Score:

  9. Features Analysis

    Correlation:

    RFE(Recursive feature elimination)

    P-value

  10. Features cleaning

  11. Model fitting and evaluation after features reduction

  12. Visualization of difference between ground truth and predicted values

  13. Cross Validation

  14. Reference

  15. Helping material