知识卡片
稀疏与异质偏好不同
交互稀疏是矩阵中可观察记录少,异质偏好则是某些用户难以与大多数人形成稳定相似关系。增加样本与寻找更合适的偏好表示分别针对不同问题;不能把所有协同失败都归因为没有足够数据。
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
Gray sheep users are those whose perceptions differ and may neither agree nor disagree with most users. Identification is problematic in CF as they may fetch issues for producing dependable, collaborative recommendations. Cold-start concerns occur when items included in the catalog have either nil or limited interactions. CF algorithms depend on item interactions for making recommendations, thus posing a problem.
参考来源
- 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第03章《深度推荐与评估指标》(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-C3.html)
- 结论依据:摘录中关于稀疏与异质偏好不同的定义、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:Gray sheep users are those whose perceptions differ and may neither agree nor disagree with most users. Identification is problematic in CF as they may fetch issues for producing dependable, collaborative recommendations. Cold-start concerns occur when items included in the catalog have either nil or limited interactions. CF algorithms depend on item interactions for making recommendations, thus posing a problem.