Machine Learning
What is the type of machine learning algorithms used to draw inferences from data without labels?
Unsupervised Learning.
Card 1 of 276. Answer: Unsupervised Learning.
Unsupervised Learning.
Card 1 of 276. Answer: Unsupervised Learning.
About · Quizzes · History of AI · To suggest new questions or report an error send me a dm.
Unsupervised learning. These algorithms see only input features, with no target column, and infer structure from the inputs themselves. In probability terms, supervised learning models p(y | x), while unsupervised learning studies p(x): where the data is dense, how it groups and which variables move together.
Association rule mining is a classic example. A shop has 1,000 unlabeled receipts: 200 contain bread and 100 contain both bread and butter, so the rule bread → butter has support 100 / 1,000 = 10% and confidence 100 / 200 = 50%. If 250 receipts contain butter, the lift is 0.5 / 0.25 = 2: bread buyers are twice as likely as average to buy butter.
Interviewers then ask how it differs from its neighbors. Semi-supervised learning uses a few labels plus many unlabeled rows. Self-supervised learning creates labels from the data itself, such as predicting a masked word. Reinforcement learning learns from rewards. In practice, unsupervised output often feeds a supervised model as cluster IDs or compressed features.