. Project
Machine Learning data Sets (Supervised and Unsupervised Learning)
Project Details:
1) Data Preparation:
Provide appropriate plots and interpretations for the attributes of the dataset. Use Ordinal, one-hot
and/or Dummy variable encoding wherever it is necessary.
Analysis should include the standalone attributes as well relationships amongst the attributes.
(10 points)
2.) Supervised Learning – Classification/Regression:
Perform Analysis using SVC/SVR, Logistic Regression/Ridge/Lasso, Random Forest
classifier/repressor, and Gradient Boosting classifier/regressor.
Select an appropriate parameter grid to search for best hyper parameters for each model while finding
the best model.
Provide the appropriate plots and interpretations.
(20 points)
4) Feature Engineering:
Using Principal Component Analysis, determine which attributes are important
for the analysis. (Do analysis of scikit-learn- permutation feature importance, scikit-learn Partial
dependence methods for features for important attributes and another using PCA for important
components).
Compare results by performing Step2.
(10 points)
5.) Unsupervised Learning:
Do the clustering techniques on the dataset using Unsupervised Learning methods 2 methods Means, DBscan. Provide the appropriate plots and Interpretations. Note: Use any two clustering
methods.
(10 points)
6) Use pipelines where appropriate for the above techniques
(10 points)

Purchase answer to see full
attachment