Data Science project on the Diabetes dataset from Kaggle, dataset can be found here. Designed and developed a E2E diabetes prediction application using data science concepts and flask. Deployment using Docker containerization and kubernetes on google cloud platform.
- Older people in the age range 35-55 have an increased chance of being diagnosed with diabetes.
- People having higher glucose content are vulnerable to diabetes diagnosis.
The Decision Tree model was the best model out of all the models used in the project. Below is the confusion matrix of the model. Metrics like PRF1 are used for evaluation.
This model is deployed on a local host using html, flask application. The initial page looks like as shown below.
We enter the required details for the prediction.
The final page after we click on Predict looks as show below.