Assessing the robustness of multi-criterial recommendation systems

Being able to comprehensively understand the intri- cacies of recommendation systems can be a significant competitive advantage. No surprise, then, that marketers are increasingly preferring robust and cost-effective multi-criteria recommenda- tion systems in this age of personalisation and lifetime value maximisation. Multi criterion recommendation systems as a methodology is already fairly well analysed. Yet, as selection criterion complexity increases - it becomes an exponentially more difficult task to check the robustness of such a system. The authors try to evaluate the strength of these systems as a part of this research. In this paper, we empirically compare two different mechanisms which measure similarities and ratings to evaluate their rating estimation performance. The results achieved can be considered one step closer to the understanding of how multidimensional evaluation can be used to assess underlying value of multi-criteria ratings.