Python 通过对两者使用相同的字符串来替换列中的相似字符串

带有以下玩具数据帧:

import pandas as pd

df = pd.DataFrame(
    [
        ("Bob Fisherman", "Income", 150, 2022),
        ("Bob Heris", "Income", 100, 2021),
        ("Bob Felony", "Income", 100, 2021),
        ("Bob Felany", "Expense", 50, 2022),
        ("Bob Haris", "Expense", 100, 2022),
        ("Bob Disherman", "Expense", 100, 2021),
    ],
    columns=["Name", "Type", "Amount", "Year"],
)
print(df)
# Output
            Name     Type  Amount  Year
0  Bob Fisherman   Income     150  2022
1      Bob Heris   Income     100  2021
2     Bob Felony   Income     100  2021
3     Bob Felany  Expense      50  2022
4      Bob Haris  Expense     100  2022
5  Bob Disherman  Expense     100  2021

以下是使用来自Python标准库difflib模块的SequenceMatcher个类的帮助来完成此操作的一种方法:

from difflib import SequenceMatcher

def similar(a, b):
    """Get similarity ratio between a and b.

    Args:
        a: value.
        b: other value.

    Returns:
        Similatity ratio.

    """
    return SequenceMatcher(None, a, b).ratio()


def replace_similar_values(df, col, value):
    """Main helper function.

    Args:
        df: target dataframe.
        col: target column.
        value: value to use for replacement.

    Returns:
        Modified dataframe.

    """
    df = df.assign(
        Match=df[col].map(
            lambda x: max(
                [similar(x, value)],
                key=lambda x: x if x != 1 else 0,
            )
        )
    )
    df["Name"] = df.apply(lambda x: value if x["Match"] >= 0.8 else x["Name"], axis=1)
    df = df.drop(columns="Match")
    return df

然后:

correct_names = ["Bob Fisherman", "Bob Felony", "Bob Haris"]

for correct_name in correct_names:
    df = replace_similar_values(df, "Name", correct_name)

print(df)
# Output
            Name     Type  Amount  Year
0  Bob Fisherman   Income     150  2022
1      Bob Haris   Income     100  2021
2     Bob Felony   Income     100  2021
3     Bob Felony  Expense      50  2022
4      Bob Haris  Expense     100  2022
5  Bob Fisherman  Expense     100  2021