Mastering OpenCV 4 with Python

Unsupervised machine learning

In unsupervised learning, there is no labeled output. In this sense, there is a collection of samples, but the corresponding output values for each sample are missing (the collection of samples has not been labeled, classified, or categorized). The goal of unsupervised learning is to model and infer the underlying structure or distribution in the collection of samples. Therefore, in unsupervised learning, the algorithm does not find out the right output, but it explores the data and can make inferences from the data trying to reveal hidden structures in it. Clustering, or dimensionality reduction, are two of the most common algorithms used in unsupervised learning.