查找DataFrame的两个字符串列之间的相似性

import pandas as pd
from difflib import SequenceMatcher
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics.pairwise import cosine_similarity

# 创建示例DataFrame
data = {'col1': ['apple', 'banana', 'orange'],
        'col2': ['aple', 'bananna', 'ornge']}
df = pd.DataFrame(data)

# 使用编辑距离计算相似性
df['edit_distance'] = df.apply(lambda row: SequenceMatcher(None, row['col1'], row['col2']).ratio(), axis=1)

# 使用余弦相似度计算相似性
vectorizer = CountVectorizer().fit_transform(df['col1'] + ' ' + df['col2'])
cosine_similarities = cosine_similarity(vectorizer)
df['cosine_similarity'] = cosine_similarities.diagonal()

# 使用Jaccard相似系数计算相似性
df['set1'] = df['col1'].apply(set)
df['set2'] = df['col2'].apply(set)
df['jaccard_similarity'] = df.apply(lambda row: len(row['set1'].intersection(row['set2'])) / len(row['set1'].union(row['set2'])), axis=1)

# 打印结果
print(df)