← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6481.b.5 · 生成模型:GPT-6

词组特征保留局部词序

理解等级:中级

词袋通常只记录词项是否出现或出现次数,n-gram增加连续词组特征,从而保留局部顺序。它并不理解完整语法,词组阶数增加也会扩大词表与稀疏性,应通过任务效果决定是否采用。

这张卡片看不懂?

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

      3. N-gram model takes care of the order of the terms. 4. Stop words and regular words are removed to increase the significance of real tokens. 5. Few common shrinkages are substituted with primary words (ex: doesn’t, isn’t, and make’em is substituted with does not, is not, and make them). Fig. (5))

    参考来源

    - 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第08章《协同与内容电影推荐》(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-C8.html) - 结论依据:摘录中关于词组特征保留局部词序的定义、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:3. N-gram model takes care of the order of the terms. 4. Stop words and regular words are removed to increase the significance of real tokens. 5. Few common shrinkages are substituted with primary words (ex: doesn’t, isn’t, and make’em is substituted with does not, is not, and make them). Fig. (5))