Assessment of Rating Prediction Techniques under Response Uncertainty

被引：1

作者：

Sizov, Sergej ^{[1
]}

机构：

[1] Heinrich Heine Univ Dusseldorf, Fac Arts & Humanities, Web Sci Grp, D-40225 Dusseldorf, Germany

来源：

PROCEEDINGS OF THE 2016 ACM WEB SCIENCE CONFERENCE (WEBSCI'16) | 2016年

关键词：

RMSE; uncertainty; collaborative filtering; recommender systems; ordinal data;

D O I：

10.1145/2908131.2908203

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

An objective assessment of collaborative filtering techniques and recommender systems requires application of suitable predictive accuracy metrics. In real life, individuals meet their decisions with considerable uncertainty. We accordingly justify underlying assumptions of quality assessment and propose and appropriate uncertainty-aware evaluation methodology for rating predictions.

引用

页码：363 / 364

页数：2