The problem is Machine Learning applied to the Prediction of Obesity and Overwei

The problem is Machine Learning applied to the Prediction of Obesity and Overweight
Identify a real-world problem of your choice. The most important requirement, apart from your good understanding of the problem, is that it would make for a good data analytics application, for example, that you can see how this problem would be more successfully tackled using data analytic techniques. Feel free to choose any of the applications you prefer (related or not to your work environment).After selecting the problem, you will investigate, and discuss the following questions:
What exactly is the business decision you want to support with this solution?
Why did you select this as a good data driven decision making problem?
What data will you use and where will you get it?
Why and how do you expect your solution will add value?
Explain all of the steps towards the deployment of your solution.
What type of data analytic task(s) do you need to perform?
What is your target variable (if any)?
What are the features?
What exactly would be your training data?
How will you evaluate model performance?
Describe the problem :The problem is Machine Learning applied to the Prediction of Obesity and Overweight

Then, describe the data that you will use to support your decision. Make sure you mention where the data is procured from. Describe the characteristics of the dataset, the variable types and what information you can get from the features.
In the context of the data mining process discussed in class, explain, step by step, how you would develop a data-driven decision-making solution to this problem. At each step, explain the various tasks you must perform, the different techniques, models etc. For every modelling task, identify your target variable and explain why it is appropriate for your objective and how its characteristics lead you to consider the particular model(s). Explain how you would perform model selection and evaluation and how you would try to address any potential problems you identify at this stage.
Please note that you are not required to perform any of the analyses discussed here, but you are not discouraged from experimenting with the data in any way you see suitable.