Models of the representation and clustering of semistructured information

A scheme for solving the problem of binary clustering of semistructured information is proposed. Different methods for the representation of the raw data of the clustering problem are examined. Methods for the progressive reduction and successive merging of clusters, as well as the initial cluster arrangement model, are considered. Estimates of the number of clusters that are required to solve the clustering problem are given. A method of binary clustering of points on a circumference is proposed.