Data clustering belongs to unsupervised learning. Unsupervised learning is a method of machine learning. It does not require manpower to input labels. It is an alternative to supervised learning and reinforcement learning. In supervised learning, the typical tasks are classification and regression analysis, and manual pre-prepared examples need to be used. Among them, the predictive regression of data can be regarded as a one-way classification, and text classification can be regarded as a purposeful clustering. Text content and URL reordering are also a process of clustering. It seems that all methods in data analysis can be converted. It is a clustering problem.
The general method of clustering is to transform features to other dimensional spaces, for example, to map their features to divisions in other spaces through Gaussian functions, such as the high-dimensional space angle division in the vector space model.