Application of Personalized Ranking Models Based on Expert Evaluations for Sorting Goods on E-commerce Web Resources

摘要

The article describes the application of a personalized model for ranking search results based on expert evaluations when ranking goods for e-commerce resources. The role of experts is played by system users, with a high rate of agreement on opinions relative to the current user. The measure of consensus is determined according to a model built using the generalized iterative algorithm of the Group Method of Data Handling GIA GMDH. This model is based on an analysis of the users’ social and personal characteristics.