推荐系统：通过surprise近邻查找相似用户和相似项目

userId,movieId,rating,timestamp
1,1,4.0,964982703
1,3,4.0,964981247
1,6,4.0,964982224
1,47,5.0,964983815
1,50,5.0,964982931
1,70,3.0,964982400
1,101,5.0,964980868
...

# -*- coding: utf-8 -*-
import pandas as pd
file_path = 'data/ratings.csv'
data_frame = pd.read_csv(file_path)
print(len(data_frame.groupby("userId")))
print(len(data_frame.groupby("movieId")))
# 610  9724

# -*- coding: utf-8 -*-
from surprise import KNNBasic, Dataset, Reader
# 数据读取
file_path = 'data/ratings.csv'
reader = Reader(line_format='user item rating timestamp', sep=',', skip_lines=1)
data = Dataset.load_from_file(file_path, reader=reader)
train_set = data.build_full_trainset()
# 训练模型：基于用户相似度
user_knn_basic = KNNBasic(k=10, min_k=3, sim_options={'user_based': True})
user_knn_basic.fit(train_set)
def get_similar_users(top_k, user_id):
    """
    找到相似用户
    Args:
        top_k(int): 相似用户数量
        user_id(str): 用户id
    Returns:
        list generator
    """
    user_inner_id = user_knn_basic.trainset.to_inner_uid(user_id)
    user_neighbors = user_knn_basic.get_neighbors(user_inner_id, k=top_k)
    user_neighbor_ids = (user_knn_basic.trainset.to_raw_uid(inner_id) for inner_id in user_neighbors)
    return user_neighbor_ids
print(list(get_similar_users(5, '610')))
# ['429', '508', '545', '53', '463']
# 训练模型：基于项目相似度
item_algo = KNNBasic(k=10, min_k=3, sim_options={'user_based': False})
# sim_options={'name': 'cosine','user_based': True} cosine/msd/pearson/pearson_baseline
item_algo.fit(train_set)
def get_similar_items(top_k, item_id):
    """
    相似项目
    Args:
        top_k(int): 相似项目数量
        item_id(str): 项目id
    Returns:
        list generator
    """
    item_inner_id = item_algo.trainset.to_inner_iid(item_id)
    item_neighbors = item_algo.get_neighbors(item_inner_id, k=top_k)
    item_neighbor_ids = (item_algo.trainset.to_raw_iid(inner_id) for inner_id in item_neighbors)
    return item_neighbor_ids
print(list(get_similar_items(5, '761')))
# ['423', '1009', '1804', '2099', '2899']

# -*- coding: utf-8 -*-
import pandas as pd
file_path = 'data/ratings.csv'
data_frame = pd.read_csv(file_path)
def get_user_movie(user_id):
    return data_frame[data_frame.userId == user_id].movieId.tolist()
user_movies = get_user_movie(610)
set_movies = set(movies)
similar_user_ids = [429, 508, 545, 53, 463]
for similar_user_id in similar_user_ids:
    movies = get_user_movie(similar_user_id)
    set_user_movies = set(user_movies)
    similar_score = len((set_user_movies & set_movies)) / len((set_user_movies | set_movies))
    print("{:.2f}".format(similar_score), end="\t")
    # 0.00  0.00  0.00  0.00  0.01