知识卡片
字符词组可捕捉拼写与语言线索
字符n-gram不依赖完整词边界,可利用局部字形组合识别语言、风格或近似拼写。它对变体的共享片段可有召回收益,但也可能把相近拼写而不同含义的文本混同;效果仍需具体标注与基准检验。
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
3.4.1 Analyzing “this” Even though, until now, we have only dealt with n-grams of word tokens, n-grams of characters can be useful too. For example, they can be used for language detection, or authorship attribution (deciding who among the set of authors wrote the document analyzed). Let’s solve a puzzle using character n-grams and the CountVectorizer class you learned about in listing 3.3. The Python
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第03章《词频与逆文档频率向量》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-3.html)
- 结论依据:摘录中关于字符词组可捕捉拼写与语言线索的定义、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:3.4.1 Analyzing “this” Even though, until now, we have only dealt with n-grams of word tokens, n-grams of characters can be useful too. For example, they can be used for language detection, or authorship attribution (deciding who among the set of authors wrote the document analyzed). Let’s solve a puzzle using character n-grams and the CountVectorizer class you learned about in listing 3.3. The Python