Important Links

• Document for Project Progress
• Dataset Used
• Website

Approaches

Approach 1 - Manomay Jamble

• Explored dataset - found many zero values in Cholesterol and one in RestingBP
• Applied decision tree and random forest on raw data
• Got an accuracy of around 85%
• Can replace Cholesterol values with mean retain info
• Another way to predict cholesterol values using other features

Approach 2 - Devanshu Dalal

• Explored dataset to find trends in data
• Preprocess data to remove unrealistic values and didone hot encoding
• Tried various models like Random Forest, Decision Tree, KNN, Logistic regression, Naive Bayes
• Got an accuracy of 90 percent on logistic regression and naive bayes
• Would work on preprocessing and improving accuracy

Approach 3 - Shreyash Deshmukh

• Explored data set and did EDA of data
• Cleaned the data by using data cleaning techniques
• Tried some machine learning models and found the appropriate model according to data
• Got an accuracy of 85.5 percent by KNN model
• Done with the creating input function for the user to input the data

Approach 4 - Vaishnavi Pingat

• Finished basic exploration and preprocessing of dataset
• Implemented Several Models with accuracies:
  • Logistic Regression : 91%
  • Decision Tree : 79%
  • Random Frorest : 89%
  • Bagging Classifier : 87%
  • Gradient Boost : 89%
  • Adaboost : 89%
  • Voting Classifier : 90%

Contributors

• Omkar Prabhune
• Shreyash Deshmukh
• Isha Deshpande
• Manomay Jamble
• Devanshu Dalal
• Vaishnavi Pingat