What type of machine learning is the spam filter on your email platform that automatically sends suspicious emails to your spam folder

Great job! Spam filters are an example of supervised learning because they are trained on emails labelled as ‘spam’ or ‘not spam’ to produce a model that will input a new email and decide whether or not it is spam.

========================================

 

Great job! OCR systems are an example of supervised learning because these systems are trained on thousands of pictures of hand-written letters that are labelled with their corresponding letters. The goal is that the OCR system can recognize hand-written letters that were not a part of the original training data.

========================================

Great job! Image classifiers are an example of supervised learning because they are trained on thousands of pictures labelled as ‘cat’ or ‘not cat’. The classifier can then look at a new, unlabelled image, and label it as ‘cat’ or ‘not cat’.

========================================

 

Great job! Such a system is an example of supervised learning because it trains on houses labelled with their retail value and then predicts the retail value of a new, unlabelled house.

========================================

 

Great job! Online recommender systems are an example of unsupervised learning because the system learns from unlabelled data. The system uses information about your profile or purchases and finds similar purchases to suggest.

========================================

Great job! An anomaly detection system is an example of unsupervised learning because the system decides how similar (or different) unlabelled data is in order to detect examples that are very unlike other examples in the data.

========================================

 

Great job! A system that plays an Atari game is an example of reinforcement learning because this system develops a policy for playing the game by playing that game many, many times (while receiving a reward signal based on the score).

========================================

 

Great job! A system that explores a maze to optimize the route to a finish line is an example of reinforcement learning because the system develops a policy for how to best complete the maze based on its exploration of the maze and the reward signal it receives when it reaches the finish line.