Unsupervised Machine Learning Classification Problems

Supervised learning can achieve extremely high performance, but they require very large and labelled data sets. Unsupervised learning systems are often less predictable, but as unlabelled data is usually more available than labelled data, they are important. For an image classifier, an unsupervised algoritm could not identify the image as a cat or a dog. Instead it would sort the training data set into various groups based on their similarity. This technique is helpful for market segmentation, image compression, et cetera.