R解析具有JSON数组的dataframe列，并转换为one-hot编码

import pandas as pd
import json
from sklearn.preprocessing import MultiLabelBinarizer

# 创建示例DataFrame
data = {
    'id': [1, 2, 3],
    'json_column': [
        '["apple", "banana"]',
        '["banana", "orange"]',
        '["apple", "orange"]'
    ]
}
df = pd.DataFrame(data)

# 解析JSON数组
df['parsed_column'] = df['json_column'].apply(json.loads)

# 使用MultiLabelBinarizer进行one-hot编码
mlb = MultiLabelBinarizer()
encoded_features = mlb.fit_transform(df['parsed_column'])

# 将编码结果转换为DataFrame
encoded_df = pd.DataFrame(encoded_features, columns=mlb.classes_)

# 合并原始数据和编码后的数据
result_df = pd.concat([df.drop(['json_column', 'parsed_column'], axis=1), encoded_df], axis=1)

print(result_df)

def safe_json_loads(s):
    try:
        return json.loads(s)
    except json.JSONDecodeError as e:
        print(f"Error parsing JSON: {e}")
        return []

df['parsed_column'] = df['json_column'].apply(safe_json_loads)