The essence of the algorithm lies in the text analysis of sites or documents, in which the site being compared is compared with a given collection. As a basic collection, we took the sites and categories of Yandex Catalog.
For each of the sites studied, a thematic vector is calculated, which is compared with vectors counted for sites from each thematic category. Subjects are determined by the closest category vector.
Note: The content field can be sent through the GET method and via the POST method.
