appendix-a
Appendix A. Solutions to the exercises
Chapter 2: Types of machine learning
For the questions in this chapter, your answers don’t need to match mine. If you have different ideas for models used in these applications, they might be great! I encourage you to look them up in the literature, and if they don’t exist, try to implement them.
Exercise 2.1
For each of the following scenarios, state if it is an example of supervised or unsupervised learning. Explain your answers. In cases of ambiguity, pick one and explain why you picked it.
- A recommendation system on a social network that recommends potential friends to a user
- A system in a news site that divides the news into topics
- The Google autocomplete feature for sentences
- A recommendation system on an online retailer that recommends to users what to buy based on their past purchasing history
- A system in a credit card company that captures fraudulent transactions
Solution
Depending on how you interpreted the problem and the dataset, each of these can be considered an example of supervised or unsupervised learning. It is completely OK (and encouraged!) to have different answers, as long as the reasoning behind them is correct.