11.11音视频点播推荐

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import linear_kernel

# 假设我们有以下音视频内容描述
videos = {
    'video1': 'A thrilling action movie with lots of explosions.',
    'video2': 'A romantic comedy about two people finding love.',
    'video3': 'An educational documentary on space exploration.'
}

# 用户偏好描述
user_preference = 'I love action movies.'

# 使用TF-IDF向量化内容描述
tfidf = TfidfVectorizer(stop_words='english')
tfidf_matrix = tfidf.fit_transform(videos.values())

# 计算用户偏好与各视频的相关性
user_tfidf = tfidf.transform([user_preference])
cosine_similarities = linear_kernel(user_tfidf, tfidf_matrix).flatten()

# 获取最相关的视频
related_videos_indices = cosine_similarities.argsort()[:-2:-1]
for idx in related_videos_indices:
    print(list(videos.keys())[idx])