知识卡片
情境建模直接扩展预测
情境建模把情境与用户、物品一同作为预测条件,而非只在输入数据或最终列表上筛选。用户—物品矩阵可扩展为含情境轴的张量。表达能力增加也带来更多稀疏组合,需要与前后过滤的质量和复杂度比较。
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
Contextual modeling: Unlike the above paradigms, contextual information is used during the recommendation phase (neither before nor after). It considers contextual data an additional entity beyond users and items to produce “multidimensional” recommendations. Predictive models (such as decision trees, regression, and probabilistic models, among others) [56], matrix (or tensor) factorization [57], heuristic calculations [58], and kernel methods [54] can also be used in this setting.
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:摘录中关于情境建模直接扩展预测的定义、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:Contextual modeling: Unlike the above paradigms, contextual information is used during the recommendation phase (neither before nor after). It considers contextual data an additional entity beyond users and items to produce “multidimensional” recommendations. Predictive models (such as decision trees, regression, and probabilistic models, among others) [56], matrix (or tensor) factorization [57], heuristic calculations [58], and kernel methods [54] can also be used in this setting.