Pandas - 根据 str 包含从另一列创建带有值的新列

Question

I have two DataFrames. One with multiple columns and other with just one. So what I need is to join based on partial str of a column. Example:

df1

| Name     |       Classification       |
| -------- | -------------------------- |
| A        | Transport/Bicycle/Mountain |
| B        | Transport/City/Bus         |
| C        | Transport/Taxi/City        |
| D        | Transport/City/Uber        |
| E        | Transport/Mountain/Jeep    |

df2

| Category |
| -------- | 
| Mountain |
| City     | 

As you can see the order on Classification column is not well difined.

Derisable Output

| Name     |       Classification       | Category  |
| -------- | -------------------------- |-----------|
| A        | Transport/Bicycle/Mountain | Mountain  |
| B        | Transport/City/Bus         | City      |
| C        | Transport/Taxi/City        | City      |
| D        | Transport/City/Uber        | City      |
| E        | Transport/Mountain/Jeep    | Mountain  |

I'm stuck on this. Any ideas?

Many thanks in advance.

Answer 1

This implementation does the trick:

def get_cat(val):
    for cat in df2['Category']:
        if cat in val:
            return cat
    return None

df['Category'] = df['Classification'].apply(get_cat)

Note: as @Justin Ezequiel pointed out in the comments, you haven't specified what to do when Mountain and City exists in the Classification. Current implementation uses the first Category that matches.

Answer 2

You can try this:

dff={"ne":[]}

for x in df1["Classification"]:
    if a in df2 and a in x:
        dff["ne"].append(a)
df1["Category"]=dff["ne"]

df1 will look like your desirable output.