METHOD AND SYSTEM FOR CLUSTERING DOCUMENTS Russian patent published in 2021 - IPC G06F16/35 

FIELD: computer technology.

SUBSTANCE: method and a system for forming clusters of documents using a generalized metric parameter are implemented. The first document and the second document are received for a potential cluster containing the first document and the second document, and the first metric parameter is determined, indicating the degree of mutual complementation of the content of documents in the potential cluster, and the second metric parameter, indicating the degree of dilution of the content of documents in this potential cluster. The generalized metric parameter is determined based on the first metric parameter and the second metric parameter. Based on the generalized metric parameter, a cluster containing the first and second documents is formed. Another document (or documents) or clusters can be added to a cluster by defining an updated generalized metric parameter for a potential cluster and comparing the updated generalized metric parameter with the generalized metric parameter.

EFFECT: increase in the efficiency of document clustering and reduction in the loss of bandwidth of communication channels.

32 cl, 6 dwg