Building a real-time recommendation system using Python and Redis: how to provide personalized recommendations

Building a real-time recommendation system using Python and Redis: How to provide personalized recommendations

Introduction:
In the era of modern information explosion, users are often faced with a large number of options and information, so personalized recommendation systems become more and more important. This article will introduce how to use Python and Redis to build a real-time personalized recommendation system, and show how to use the powerful functions of Redis to provide personalized recommendations.

1. What is a personalized recommendation system
A personalized recommendation system is based on the user's interests and behavior, combined with algorithms and machine learning technology, to recommend content or products that suit the user's interests and needs. The core of the personalized recommendation system is to analyze and understand the user's behavior and interests, so as to accurately predict the user's preferences and needs and provide corresponding recommended content.

2. Introduction to Redis
Redis is an open source in-memory database with efficient reading and writing speed and rich data structure support. It can be used in a variety of scenarios such as caching, message queues, and real-time counters. In the personalized recommendation system, Redis can be used as a storage and analysis tool for user behavior and interests, providing real-time data support for the recommendation system.

3. Basic environment construction
Before building the real-time recommendation system, we need to install and configure the Python and Redis environments.

1. Install Python and the corresponding dependent libraries
   Enter the following commands on the command line to install Python and the dependent libraries:

   $ sudo apt-get update
   $ sudo apt-get install python3 python3-pip
   $ pip3 install redis

2. Install Redis
   Enter the following command on the command line to install Redis:

   $ sudo apt-get install redis-server

4. Real-time recommendation system design
This article will take the "Movie Recommendation System" as an example to show how to use Python Build a real-time personalized recommendation system with Redis.

1. Data preprocessing
   First, we need to prepare some movie data, including the name, classification, rating and other information of the movie. Store these data in Redis to facilitate subsequent data query and recommendation.

import redis

# 连接Redis
r = redis.Redis(host='localhost', port=6379)

# 存储电影数据
movies = [
    {"id": 1, "title": "电影1", "category": "喜剧", "rating": 4.5},
    {"id": 2, "title": "电影2", "category": "动作", "rating": 3.8},
    {"id": 3, "title": "电影3", "category": "爱情", "rating": 4.2},
    # 添加更多电影数据...
]

for movie in movies:
    r.hmset("movie:%s" % movie["id"], movie)

2. User Behavior Analysis
   Next, we need to collect users’ ratings or viewing records of movies and store them in Redis for subsequent personalized recommendations.

# 添加用户行为数据
user1 = {"id": 1, "ratings": {"1": 5, "2": 4, "3": 3}}
user2 = {"id": 2, "ratings": {"1": 4, "2": 3, "3": 2}}
user3 = {"id": 3, "ratings": {"2": 5, "3": 4}}
# 添加更多用户数据...

for user in [user1, user2, user3]:
    for movie_id, rating in user['ratings'].items():
        r.zadd("user:%s:ratings" % user["id"], {movie_id: rating})

3. Personalized recommendation
   Finally, we use a personalized recommendation algorithm based on the collaborative filtering algorithm to recommend users.

# 获取用户的观看记录
def get_user_ratings(user_id):
    return r.zrange("user:%s:ratings" % user_id, 0, -1, withscores=True)

# 获取电影的评分
def get_movie_rating(movie_id):
    movie = r.hgetall("movie:%s" % movie_id)
    return float(movie[b"rating"])

# 个性化推荐算法
def personalized_recommendation(user_id, top_n=3):
    user_ratings = get_user_ratings(user_id)
    recommendations = []

    for movie_id, rating in user_ratings:
        related_movies = r.smembers("movie:%s:related_movies" % movie_id)
        for movie in related_movies:
            if r.zrank("user:%s:ratings" % user_id, movie) is None:
                recommendations.append((movie, get_movie_rating(movie)))

    return sorted(recommendations, key=lambda x: x[1], reverse=True)[:top_n]

# 输出个性化推荐结果
user_id = 1
recommendations = personalized_recommendation(user_id)
for movie_id, rating in recommendations:
    movie = r.hgetall("movie:%s" % movie_id)
    print("电影：%s, 推荐评分：%s" % (movie[b"title"], rating))

5. Summary
This article introduces how to use Python and Redis to build a real-time personalized recommendation system. Through the powerful functions of Redis, we can easily store and analyze user behavior and interests, and provide users with personalized recommendation content. Of course, this is only the basis of a personalized recommendation system. More complex algorithms and technologies can be applied according to actual needs to improve the recommendation effect. In practical applications, issues such as data security and performance also need to be considered, but this article provides a simple example that I hope will be helpful to readers.

The above is the detailed content of Building a real-time recommendation system using Python and Redis: how to provide personalized recommendations. For more information, please follow other related articles on the PHP Chinese website!