how to select rows that contains one or more keywords(from an existing list) in pandas?

user_12278890 0
Viewed 786

I'm trying to select every row of a pandas.DataFrame which df['Title'] has one(or more) of the keywords elements.

consider this list as keywords:
keywords = ['k_1', 'k_2', 'k_3', 'k_4']

I've tried this approach which did not worked out for me:
df[df['Title'].str.contains(keywords)]

pandas dataframe python
2 Answers
df[df["Title"].apply(lambda x: any(k in x for k in keywords))]
user_118039450

Create a regex pattern and use str.findall:

Setup:

df = pd.DataFrame({'Title': ['k_1 and k_2', 'k_3 alone', 'k_z not here']})
keywords = ['k_1', 'k_2', 'k_3', 'k_4']
pattern = fr"\b({'|'.join(keywords)})\b"

df['Keywords'] = df['Title'].str.findall(pattern)

Output:

>>> df
          Title    Keywords
0   k_1 and k_2  [k_1, k_2]
1     k_3 alone       [k_3]
2  k_z not here          []

>>> print(pattern)
\b(k_1|k_2|k_3|k_4)\b

Get rows:

>>> df[df['Title'].str.findall(pattern).astype(bool)]
         Title
0  k_1 and k_2
1    k_3 alone
user_152399510
Related