Data Management

Explanations of Recommendations Considering Human Factors in Recommender Systems

Download Now Free registration required

Executive Summary

Traditional evaluation metrics based on statistical formulas employed to assess the performance of a recommender system are now considered to be inadequate when utilized solely. New metrics considering the quality of user-system interaction alongside with traditional ones have been proposed in evaluation process in order to arrive at more adequate results. Generating explanations for recommendations is a research topic that has emerged as a way to evaluate the system with respect to various criteria considering users' opinions and feelings. This paper presents the state-of-the-art with respect to the explanation of a recommendation.

  • Format: PDF
  • Size: 472.32 KB