Clustering algorithms are designed to split a dataset up into groups. Once trained, any new data can be assigned to a group when it arrives. Suppose you are working with a dataset of customer information for an e-commerce store. You might use clustering to identify groups of customers, for example, business/private customers. This information can then be used to make decisions about how to best serve those customer types.
You might also use clustering as a preparatory step before applying supervised learning. For example, a dataset of images may require manual labeling, which is often time-consuming and costly. If you can segment the dataset into groups with a clustering algorithm, then you may be able to save time by only labeling a fraction of the images, and then assuming that each cluster contains images with the same label.
Clustering has also been applied to computer...