双11AI变脸推荐

import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import linear_kernel

# 假设有商品数据和用户历史购买数据
products = pd.DataFrame({
    'product_id': [1, 2, 3],
    'name': ['商品 A', '商品 B', '商品 C'],
    'description': ['这是商品 A 的描述', '这是商品 B 的描述', '这是商品 C 的描述']
})

user_history = pd.DataFrame({
    'user_id': [1],
    'product_id': [1]
})

# 将商品描述转换为 TF-IDF 向量
tfidf = TfidfVectorizer(stop_words='english')
products['description'] = products['description'].fillna('')
tfidf_matrix = tfidf.fit_transform(products['description'])

# 计算相似度
cosine_sim = linear_kernel(tfidf_matrix, tfidf_matrix)

# 根据用户历史购买的商品推荐相似商品
def get_recommendations(product_id):
    idx = products[products['product_id'] == product_id].index[0]
    sim_scores = list(enumerate(cosine_sim[idx]))
    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)
    sim_scores = sim_scores[1:3]  # 取前两个最相似的商品
    product_indices = [i[0] for i in sim_scores]
    return products['product_id'].iloc[product_indices]

recommended_products = get_recommendations(user_history['product_id'].iloc[0])
print(recommended_products)