我需要的是在某个“单词距离”内找到单词的功能。 “包”和“工具”这个词在一句话中很有意思“他车上装了一袋工具。”
使用Quanteda kwic功能,我可以单独找到“包”和“工具”,但这通常会让我产生过多的结果。我需要例如'bag'和'tools'在五个单词之内。
您可以使用fcm()
函数计算固定窗口内的共现,例如5个单词。这创建了“特征共现矩阵”,并且可以为任何大小的令牌范围或整个文档的上下文定义。
对于您的示例,或者至少基于我对您的问题的解释的示例,这将看起来像:
library("quanteda")
## Package version: 1.4.3
## Parallel computing: 2 of 12 threads used.
## See https://quanteda.io for tutorials and examples.
txt <- c(
d1 = "He had a bag of tools in his car",
d2 = "bag other other other other tools other"
)
fcm(txt, context = "window", window = 5)
## Feature co-occurrence matrix of: 10 by 10 features.
## 10 x 10 sparse Matrix of class "fcm"
## features
## features He had a bag of tools in his car other
## He 0 1 1 1 1 1 0 0 0 0
## had 0 0 1 1 1 1 1 0 0 0
## a 0 0 0 1 1 1 1 1 0 0
## bag 0 0 0 0 1 2 1 1 1 4
## of 0 0 0 0 0 1 1 1 1 0
## tools 0 0 0 0 0 0 1 1 1 5
## in 0 0 0 0 0 0 0 1 1 0
## his 0 0 0 0 0 0 0 0 1 0
## car 0 0 0 0 0 0 0 0 0 0
## other 0 0 0 0 0 0 0 0 0 10
这里,术语包在第一个文档中的工具的5个标记内发生一次。在第二个文件中,它们相距超过5个令牌,因此不计算在内。