如何在Pandas DataFrame中根据指定列的关键词筛选并返回另一列的对应值

Hey there! Let's walk through how to tackle both of your pandas needs clearly and simply.

1. Basic: Extract corresponding names for a single keyword #

You already got the first part right with str.contains to create a boolean mask—now we just need to use that mask to slice the DataFrame and pull out the names column.

First, let's set up your sample DataFrame:

import pandas as pd

df = pd.DataFrame({
    'names': ['a', 'b', 'c'],
    'words': ['apple', 'apple', 'pear']
})

Then, apply your mask and extract the values:

# Create the boolean mask for rows where 'words' contains 'apple'
mask = df['words'].str.contains('apple')

# Use the mask to filter the DataFrame and get the 'names' column, then convert to a list
matching_names = df.loc[mask, 'names'].tolist()

print(matching_names)  # Output: ['a', 'b']

If you need exact matches (not partial contains, e.g., avoiding matches for 'applepie'), replace str.contains with a direct equality check:

mask = df['words'] == 'apple'

2. Advanced: Map multiple keywords to their corresponding names #

For your second request—getting independent names lists for each keyword—you can use a dictionary comprehension to loop through your keywords and build a map of keyword-to-names.

Example with multiple keywords:

target_keywords = ['apple', 'pear']

# Build a dictionary where each key is a keyword, value is the list of matching names
keyword_name_map = {
    keyword: df.loc[df['words'].str.contains(keyword), 'names'].tolist()
    for keyword in target_keywords
}

print(keyword_name_map)
# Output: {'apple': ['a', 'b'], 'pear': ['c']}

Again, swap str.contains with == if you need exact matches. You can also add case=False to str.contains if you want case-insensitive matching (e.g., matching 'Apple' or 'APPLE' too):

mask = df['words'].str.contains('apple', case=False)

内容的提问来源于stack exchange，提问作者Gazoo