Diabetes is one of the major health issues which is in the present world. There are many people who dont know if they are suffering for diabetes and this makes the situatuion more critical. There are some places where people cant get necessary infrastructure to test. So I designed this model of course this model requires several complex features. This is just a basic solution to prove that even health issues can be solved or predicted using data science.
Today this work may look small but for better tomorrow this is the brighter step. This shows solutions to several problems that can be solved using various Machine Learning models.
I did this project in 2 ways
- Using simple Decision tree Classifier.
- Using Artificial Neural Network
The accuracy in predicing of model is around 78% and to be frank this does not look that effective. We took only limited features to predicted output and the dataset has only 700 tupeles. If we can get more data then the accuray of this model can be incresed.
You can get the required dataset and the source code for both models in this repo