Question 4. Consider the following hypothetical scenario. A car company would like to use a Bayesian Network model to better predict whether a certain customer will buy a specific car, so they can focus their efforts on developing certain car models. Specifically, they want to label pairs of customers and car models according to whether they belong to the target class 'buys'. The manufacturer has selected eight attributes, each taking values from {yes, no}, namely Basic features of the car: - '5-star safety rating': whether the car model has been awarded with the highest safety rating (5-stars in this case) - 'side-airbags': whether the car model includes side airbags 'large engine capacity': whether the car has a capacity of at least 2 litres 'expensive' whether the car is expensive Characteristics of the client: 'young': whether the client is young 'rich': whether the client is rich 'family': whether the client is a family 'interested': whether the client is interested in the car 3 Your task is to provide a fully specified Bayesian network based on the following brief: 1. The Bayesian network must use all attributes provided by the agency. 2. You may include as many new attributes as you like, as long as they add valuable information and it is realistic for the company to have access to them. 3. Whether a car is expensive has to depend on all basic features the car has, and it should be easy to incorporate additional such features into the Bayesian network. 4. As not all customers are equally likely to be interested in the same car, the Bayesian network must explicitly model what makes this customer interested in this car, i.e., it has to relate information about the car to information about the customer. Justify your design choices and explain how additional features of a car would be included in your model, using at most one page of text (approximately 500 words)