Affiliation:
1. University of Duisburg-Essen, Germany
Abstract
Most IR experiments lack both internal and external validity. Top performance is mostly an illusion, given the lack of solid statistical evidence. Thus, reviewers should ignore performance numbers in their judgement, and pay more attention to proper methodology. Researchers need to focus more on performance analysis and prediction.
Publisher
Association for Computing Machinery (ACM)
Subject
Hardware and Architecture,Management Information Systems
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献