python - 如何使用 Pandas python 中的另一个数据框替换数据框中的单词

Question

我有两个数据框:

df:

id   string_data
1    My name is Jeff
2    Hello, I am John
3    I like Brad he is cool.

另一个名为 allnames 的数据框包含如下名称列表:

id  name
1   Jeff
2   Brad
3   John
4   Emily
5   Ross

我想用 "Firstname" 替换 df 中出现在 allnames['name'] 中的所有单词

预期输出:

id   string_data
1    My name is Firstname
2    Hello, I am Firstname
3    I like Firstname he is cool.

我试过这个:

nameList = '|'.join(allnames['name'])
df['string_data'].str.replace(nameList, "FirstName", case = False))

但是它几乎替换了99%的词

Answer 1

如果将单词边界添加到 Series.str.replace，您的解决方案应该有效:

nameList = '|'.join(r"\b{}\b".format(x) for x in allnames['name'])
df['string_data'] = df['string_data'].str.replace(nameList, "FirstName", case = False)
print (df)
   id                   string_data
0   1          My name is FirstName
1   2         Hello, I am FirstName
2   3  I like FirstName he is cool.

或者通过字典用get和join替换值:

d = dict.fromkeys(allnames['name'], 'Firstname')
f = lambda x: ' '.join(d.get(y, y) for y in x.split())
df['string_data'] = df['string_data'].apply(f)
print (df)
   id                   string_data
0   1          My name is Firstname
1   2         Hello, I am Firstname
2   3  I like Firstname he is cool.

编辑:您可以通过 lower 将所有值转换为小写:

d = dict.fromkeys([x.lower() for x in allnames['name']], 'Firstname')
f = lambda x: ' '.join(d.get(y.lower(), y) for y in x.split())
df['string_data'] = df['string_data'].apply(f)