12.12视频智能分类推荐

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

# 假设我们有一个视频数据集，每个视频有一个描述字段
videos = [
    {"id": 1, "description": "科幻电影，讲述未来世界..."},
    {"id": 2, "description": "喜剧片，轻松幽默..."},
    # ...更多视频
]

# 使用TF-IDF向量化视频描述
tfidf = TfidfVectorizer(stop_words='english')
tfidf_matrix = tfidf.fit_transform(video['description'] for video in videos)

# 计算视频之间的相似度
cosine_sim = linear_kernel(tfidf_matrix, tfidf_matrix)

# 推荐函数
def get_recommendations(video_id, cosine_sim=cosine_sim):
    idx = next((i for i, v in enumerate(videos) if v['id'] == video_id), None)
    if idx is None:
        return []
    
    sim_scores = list(enumerate(cosine_sim[idx]))
    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)
    sim_scores = sim_scores[1:6]  # 获取最相似的5个视频索引
    video_indices = [i[0] for i in sim_scores]
    return [videos[i] for i in video_indices]

# 示例调用
recommended_videos = get_recommendations(1)
for video in recommended_videos:
    print(video['description'])