Machine Learning
What is unsupervised learning?
Unsupervised learning aims to detect patterns in data where no labels are given.
Card 1 of 259. Answer: Unsupervised learning aims to detect patterns in data where no labels are given.
Unsupervised learning aims to detect patterns in data where no labels are given.
Card 1 of 259. Answer: Unsupervised learning aims to detect patterns in data where no labels are given.
About · Quizzes · History of AI · To suggest new questions or report an error send me a dm.
Unsupervised learning is machine learning on data without labels: the algorithm sees only the inputs, with no target to predict, and must find structure in them itself. Rather than mapping inputs to a known output, it describes the inputs: which points are alike, which directions carry most of the variation, which points are unusual.
Its main tasks are clustering, dimensionality reduction, density estimation, anomaly detection and association rules. Example: a bank has a million card transactions and no fraud labels. An isolation forest scores each transaction by how few random splits it takes to separate it from the rest, and the 1,000 most isolated go to an analyst.
The usual follow-up is evaluation. With no ground truth there is no accuracy, so you judge results by internal scores such as the silhouette, by whether they help a downstream model, or by expert review. Be careful: k-means returns k clusters even from random noise, and your choice of features, scaling and distance shapes what gets found.