知识卡片
过度专门化限制探索
内容推荐持续围绕既有内容特征寻找相似对象,可能使结果集中在狭窄范围,难以发现用户可能喜欢的新类型。这与内容特征缺失、用户历史不足是不同限制;更准确描述旧偏好不必然扩大有用选择。
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
Challenges faced by CB include the limited association of content with the items, cold-start issues, over-specialization, and higher time consumption. Limited content refers to the scarcity of features to analyze a document, over-specialization refers to similar documents with different titles, and cold-start refers to the shortage of information concerning users and items.
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:摘录中关于过度专门化限制探索的定义、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:Challenges faced by CB include the limited association of content with the items, cold-start issues, over-specialization, and higher time consumption. Limited content refers to the scarcity of features to analyze a document, over-specialization refers to similar documents with different titles, and cold-start refers to the shortage of information concerning users and items.