movielens数据集基于协同过滤算法推荐python实现
时间: 2023-08-01 09:07:20 浏览: 99
使用Python实现基于Movielens数据集的协同过滤算法推荐,可以按照以下步骤进行:
1. 下载Movielens数据集,并将其转换为pandas DataFrame格式:
```python
import pandas as pd
# 读取数据集
ratings_data = pd.read_csv('ratings.csv')
movies_data = pd.read_csv('movies.csv')
# 数据预处理
ratings_data = ratings_data.drop(['timestamp'], axis=1)
movies_data = movies_data.drop(['genres'], axis=1)
# 合并数据
movie_ratings_data = pd.merge(ratings_data, movies_data, on='movieId')
```
2. 使用scikit-learn库的train_test_split函数将数据集划分为训练集和测试集:
```python
from sklearn.model_selection import train_test_split
train_data, test_data = train_test_split(movie_ratings_data, test_size=0.2)
```
3. 计算电影之间的相似度:
```python
# 构建电影评分矩阵
movie_ratings_matrix = train_data.pivot_table(index='userId', columns='title', values='rating')
# 计算电影之间的相似度
from sklearn.metrics.pairwise import cosine_similarity
movie_similarity_matrix = pd.DataFrame(cosine_similarity(movie_ratings_matrix.fillna(0)), index=movie_ratings_matrix.index, columns=movie_ratings_matrix.index)
```
4. 使用相似度矩阵为用户推荐电影:
```python
# 定义推荐函数
def recommend_movies(user_id, top_n):
# 获取用户没有评过分的电影
user_unrated_movies = movie_ratings_data[movie_ratings_data['userId'] == user_id][['movieId', 'title']].drop_duplicates()
user_rated_movies = train_data[train_data['userId'] == user_id][['movieId', 'title']]
user_unrated_movies = pd.merge(user_unrated_movies, user_rated_movies, on=['movieId', 'title'], how='left', indicator=True)
user_unrated_movies = user_unrated_movies[user_unrated_movies['_merge'] == 'left_only'][['movieId', 'title']]
# 计算每部电影的预测评分
movie_ratings = pd.DataFrame(columns=['movieId', 'predicted_rating'])
for movie_id in user_unrated_movies['movieId']:
similar_movies = movie_similarity_matrix[movie_id].sort_values(ascending=False)[1:6]
similar_movie_ratings = train_data[train_data['movieId'].isin(similar_movies.index)]
similar_movie_ratings = similar_movie_ratings.groupby(['movieId'])['rating'].mean()
predicted_rating = (similar_movie_ratings * similar_movies).sum() / similar_movies.sum()
movie_ratings = movie_ratings.append({'movieId': movie_id, 'predicted_rating': predicted_rating}, ignore_index=True)
# 根据预测评分为用户推荐电影
recommended_movies = pd.merge(movie_ratings, movies_data, on='movieId', how='left')
recommended_movies = recommended_movies.sort_values('predicted_rating', ascending=False).head(top_n)
return recommended_movies[['title', 'predicted_rating']]
```
以上就是基于Movielens数据集的协同过滤算法推荐的Python实现。
相关推荐
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![pdf](https://img-home.csdnimg.cn/images/20210720083512.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)
![](https://csdnimg.cn/download_wenku/file_type_ask_c1.png)