知识卡片
离线分数与在线效用有别
离线实验利用历史记录检验预测或排序,在线实验观察真实用户在监测条件下的体验和行为。历史偏好匹配不直接证明新推荐有用;还应关注可用性、多样性与新颖性。在线比较也要明确人群和目标,不能以上线本身替代验证。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- · 查看原书依据
Online experiments require a group of users to adopt RS on a monitored engine and provide information related to their experience with it. A comparison of evaluation approaches on the cross-domain recommendation's main goals noticed that most works adopt offline experiments [7].
参考来源
- 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第04章《多领域情境推荐的张量分解》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications/_epub-src/OEBPS/Text/9789815136746-C4.html)
- 结论依据:摘录中关于离线分数与在线效用有别的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:Online experiments require a group of users to adopt RS on a monitored engine and provide information related to their experience with it. A comparison of evaluation approaches on the cross-domain recommendation's main goals noticed that most works adopt offline experiments [7].