双11内容识别推荐

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

# 示例数据
data = {
    'item_id': [1, 2, 3],
    'description': ['laptop with 16GB RAM', 'smartphone with 128GB storage', 'tablet with 8GB RAM']
}
df = pd.DataFrame(data)

# 使用TF-IDF向量化文本描述
tfidf = TfidfVectorizer(stop_words='english')
df['description'] = df['description'].fillna('')
tfidf_matrix = tfidf.fit_transform(df['description'])

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

# 推荐函数
def get_recommendations(item_id, cosine_sim=cosine_sim):
    idx = df[df['item_id'] == item_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]  # 获取最相似的两个项目
    item_indices = [i[0] for i in sim_scores]
    return df['item_id'].iloc[item_indices]

# 示例调用
print(get_recommendations(1))  # 输出与item_id为1的项目最相似的两个项目的ID