Normalize Inconsistent Boolean Flags in Pandas

Question

Given a column with dirty boolean values like `['yes', 'Y', '1', 'no', 'N', '0', None]`, normalize them into pure `True`, `False`, and `pd.NA` boolean dtypes.

Accepted Answer

Optimal Solution:
import pandas as pd

bool_map = {
    'yes': True, 'y': True, '1': True, 'true': True,
    'no': False, 'n': False, '0': False, 'false': False
}

df['clean_subscribed'] = (
    df['subscribed']
    .astype(str)
    .str.strip()
    .str.lower()
    .map(bool_map)
    .astype('boolean') # Nullable boolean dtype
)

result_df = df[['clean_subscribed']]

Technical Explanation:
Cleaning the string, mapping against a canonical dictionary, and casting to `'boolean'` supports 3-valued logic (True, False, pd.NA).

subscribed
YES
0
N

clean_subscribed
true
false
false