Minimization of Product Utility Estimation Errors in Recommender Result Set Evaluations
WI-IAT '09 Proceedings of the 2009 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Volume 01
Minimization of decoy effects in recommender result sets
Web Intelligence and Agent Systems
Hi-index | 0.00 |
Recommender systems help customers toidentify products which fulfill certain requirements.Many different approaches have been developed andwork well for different domains. What has been ignoredso far is the fact that psychological phenomena can resultin suboptimal decision taking on recommender resultpages. This paper presents a user study investigating theAsymmetric Dominance Effect and the CompromiseEffect in the domain of financial services. The empiricalfindings are compared with the calculations of the SimpleDominance Model which serves as a model foridentification and explanation of such effects.Furthermore, we show how to identify misleading effectson recommender result pages, and how to correctproduct perception such that the felt utility reflects theutility calculated by recommender systems.