在单元格中排序字符串

这不是一个简单的问题。。。我建议分开,分类和加入 map :

df['states'] = df['states'].map(lambda x: ', '.join(sorted(x.split(', '))))
df

   individual                  states
0           1          Alaska, Hawaii
1           2          Alaska, Hawaii
2           3  Iowa, Kansas, Maryland
3           4      New Jersey, Newada
4           5      New Jersey, Newada

get_dummies 然后 dot

s = df.states.str.get_dummies(', ')
s.dot(s.columns+',').str[:-1]
Out[861]: 
0           Alaska,Hawaii
1           Alaska,Hawaii
2    Iowa,Kansas,Maryland
3       New Jersey,Newada
4       New Jersey,Newada
dtype: object

df['state'] = s.dot(s.columns+',').str[:-1]