Developing hybrid recommendation systems: ukrainian dimension

The lack of a consistent strategy to product recommendations, as well as the range of effective techniques to offering suggestions, are reflected in the variety of current recommendation systems. The study presents the original recommendation system, which delivers suggestions based on content and collaborative techniques and addresses the major issues in this field. The focus of the paper is hybrid recommendation systems in e-commerce on the market with a low level of implementing recommendation systems techniques. The market of recommendation systems in Ukraine, their main features are analysed. The methodology to developing hybrid recommendation systems that is relevant to the needs of Ukrainian e-commerce market is proposed. The hybrid recommendation system includes recommendation systems in four categories: Personalized recommendation, Best buy, News, Recommendation according to the survey. The alternative approach to product evaluation in proposed recommendation systems based on a combination of Wilson, Bayes, and Hacker methods is used. It is shown that this approach can be successful for recommendation systems in Ukraine. The concept's utility for users is the creation of more customised recommendations that are more attractive to them, taking into account a broader set of variables, for example, the time of publishing, the percentage of favourable comments, and personal preferences.