如何在Pandas中基于不同列生成品牌与列的关联统计表？

Got it, let's break this down to get the brand count per column you're looking for. First, let's recap your original data to make sure we're on the same page:

Original User Preference Data #

Final Brand Count by Column #

Here's the aggregated count of each brand across the three columns, exactly as you requested:

How to Calculate This (Manual + Automated) #

Manual Method

If you're doing this by hand, just tally each brand's appearance in every column:

• BMW: Shows up once in My_Fav (User0), twice in Bro_Fav (UserC, UserD), and twice in Friend_Fav (User0, UserD)
• VW: Appears twice in My_Fav (UserA, UserD), once in Bro_Fav (User0), and once in Friend_Fav (UserB)
• Honda: One occurrence in each column (My_Fav: UserB, Bro_Fav: UserB, Friend_Fav: UserA)
• Mercedes: One occurrence in each column (My_Fav: UserC, Bro_Fav: UserA, Friend_Fav: UserC)

Automated Method (For Larger Datasets)

If you have more data later, using Python's pandas library will save you time. Here's a quick script to generate the same result:

import pandas as pd

# Load your data into a DataFrame
user_data = {
    'Users_id': ['User0', 'UserA', 'UserB', 'UserC', 'UserD'],
    'My_Fav': ['BMW', 'VW', 'Honda', 'Mercedes', 'VW'],
    'Bro_Fav': ['VW', 'Mercedes', 'Honda', 'BMW', 'BMW'],
    'Friend_Fav': ['BMW', 'Honda', 'VW', 'Mercedes', 'BMW']
}
df = pd.DataFrame(user_data)

# Calculate counts for each column
brand_counts = pd.DataFrame({
    column: df[column].value_counts()
    for column in ['My_Fav', 'Bro_Fav', 'Friend_Fav']
}).fillna(0).astype(int)

# Print or display the result
print(brand_counts)

Running this script will output the exact count table you need—no manual tallying required!

内容的提问来源于stack exchange，提问作者s_khan92