Inspiration
Wanted to understand general health risks with machine learning
What it does
Person answers a typora type questionnaire about their demographics and lifestyle choices to assess risk of heart disease.
How we built it
Used Python as the backend to use the KNN model to classify the factors that could increase someone's risk of heart disease; used React, JavaScript, HTML, and CSS to create the user interface for the questionnaire.
Challenges we ran into
Original model we used was Logistic Regression, but that model overfitted by a lot by giving a 100% accuracy (no Statistical model should be 100% correct because it doesn't take account of the small percentage of chance of being incorrect), so we switched to the KNN algorithm and that gave a 91% accuracy. Also, we had the largest challenge of trying to integrate the backend and frontend code together to make the application full complete due to time constraints.
Accomplishments that we're proud of
We finished both the backend and frontend components individually for the final product.
What we learned
We learned how Machine Learning can help automate the process for survey completion.
What's next for Health Survey for Heart Disease Risk
If we have more time, we can try merge the backend and frontend code together to have a fully working product.
Log in or sign up for Devpost to join the conversation.