Developing a recommended set of movies for online movie theater customers

Abstract

Purpose: the article is devoted to the improvement of the methodology for forming a list of new films ordered by the interests of any particular client of an online cinema based on previously received expressions of interests of all clients for existing films. Discussion: сompiling a set of films based on interests requires the use of mathematical and instrumental methods that take into account the features of metrics to determine the similarity of rows in matrices of unstructured preferences of large dimension. Therefore, it is relevant to search in the field of Big Data for new solutions and their combinations. Results: it is proposed to use the properties of Pareto sets to select clients who are close in interests and to rank them by proximity to the interests of each client using evaluation vectors. The method of solving an informal problem of large dimension developed in the article is distinguished by its originality. It will automate the preparation of recommendations and increase their objectivity and accuracy. In addition, the use of Big Data technology will ensure the self-adaptation of the proposed tool to the changing interests of customers and the ability to explain the advice provided.