2021 AI Invasion: Building Medical Insurance model

AI has come to stay, it is therefore important that everybody try to learn and understand how AI is changing our world and work, it is for this reason Data Science Nigeria came up with the idea of AI Invasion where I serve as a Lead Tutor / Facilitator for NITDA, UNILAG centre. You can read more on AI Invasion here.

During this free five days program, we talked about fundamental concept of AI, Machine learning and its types, business cases and different algorithm. We worked on classification problem as well as Regression problem. In this article, I will work you through the model building for the classification problem.

We are going to be working with Medical Insurance data. Our goal is to be able to predict if a customer will purchase / subscribe to medical insurance service for a company. You can access the data here. Please, I assume you have a little background in python.

If you check the two models, you can deduce that KNearest Neigbors tends to perform better than Support Vector Machine. Lastly, you will need to also understand how to deploy your model so that a larger crowd can assess it. You can read the deployment here. Depolyment is the last stage of machine learning, however, I see many data scientist not taking it serious. Once again, to see the deployment of our model, check here.

You can get the full code used here.

PS: Some data preprocessing like treating class imbalance, etc were omitted because I assume the students during the program are total beginners. You can check my tutorial on how to increase your model.

I have a little request, if you have learnt anything here, please share to benefit others, like the article, and share your opinion. You can connect with me: Linkedin.