Objective Identify presence of heart disease.

Dataset Data: https://www.kaggle.com/ronitf/heart-disease-uci

1. What is the primary objective that you are addressing and why?

2. For all the independent variables given, state the discrete variables and their types (For example Nominal/Ordinal ).

3. Find the independent variables which are good candidates for data transformation

4. Transform any one variable which is interval/ratio to categorical variable

5. Transform any one variable which is interval/ratio to ordinal variable

6. Is there any association between any two interval/ratio variable. How will you go about it?

7. Which graph type would you prefer to use for visualizing each variable

8.What will be your criteria to select the classification algorithm?

9. How will you evaluate the performance of two classifier that you investigated and state which is better

10.Can you deploy your model in real time ? What are the challenges?

Please follow the given format below

Overview

• Objective
• Methodology

Dataset

• How many features
• Size of the dataset
• Multiple files
• What kind of data numerical or character
• Balanced or imbalanced what is the distribution
• Distribution of Training set, validation set, testing set
• Missing data and Preprocessing challenges

Feature Engineering Techniques

• Features removed
• Feature creation
• Feature ranking
• Class imbalance treatment
• Any other

Methodology

• The 3 classifiers used
• Ensemble pipeline
• Other models considered
• Hyper-parameter tuning

Results

• Table for the evaluation metric for each ML technique used
• Plot of the curves
• Conclusion