知识卡片
合并矩阵仍需实体连接
传统单域协同算法可在跨域合并后的交互矩阵上工作,但合并需要用户或物品等连接信息。两个完全分离的评分块不会因放进同一张表就自动产生可比较邻居;应明确跨块关系来自哪里。
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
there is overlap between items or/and users; CD suggestions could be produced by combining user-rating matrices from multiple domains using standard single-domain collaborative filtering algorithms (Aggregating knowledge). So, these algorithms could also be adopted for making cross-domain suggestions, as shown by Véras et al. [15], considering the CD-CARS formalization extended in this work. In the following, we briefly describe some traditional algorithms that can be
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:摘录中关于合并矩阵仍需实体连接的定义、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:there is overlap between items or/and users; CD suggestions could be produced by combining user-rating matrices from multiple domains using standard single-domain collaborative filtering algorithms (Aggregating knowledge). So, these algorithms could also be adopted for making cross-domain suggestions, as shown by Véras et al. [15], considering the CD-CARS formalization extended in this work. In the following, we briefly describe some traditional algorithms that can be