The rapid development of Internet technology has brought us into the era of information explosion. The simultaneous presentation of massive amounts of information makes it difficult for users to find the parts they are interested in, and on the other hand, makes a lot of information that few people care about become a part of the network. Such advance technology cannot be obtained by ordinary users.
The personalized recommendation system establishes the binary relationship between users and information products, and uses the existing selection process or similarity relationship to mine the potentially interesting objects of each user. The essence of personalized recommendation is information filtering. Personalized recommendation system not only has important application value in social economy, but also is a scientific problem worthy of research.
In fact, it is currently the most effective tool to solve the problem of information overload. According to the different recommendation algorithms, we will introduce the collaborative filtering system, the content-based recommendation system, the hybrid recommendation system, and the recently emerging recommendation system based on the user-product bipartite graph network structure. Combined with the characteristics and existence of these recommendation systems The deficiencies will be proposed, and be improved for possible future research.
The research of recommendation system has attracted the attention of information science, computational mathematics, statistical physics, cognitive science and other disciplines, and it is closely related to management science, consumer behavior and other researches It is also closely related. It can provide reference for researchers in different disciplines to study the recommendation system, and help Chinese scholars to understand the main progress in this field.
With the rapid development of the Internet, the number of servers connected to the Internet and the number of web pages have shown an exponential growth trend. The rapid development of Internet technology has made a large amount of information presented to us at the same time, for example, Netflix There are tens of thousands of movies on Amazon, millions of books on Amazon, and more than billions of web pages on Google, so much information, finding the part that interests you, even browsing it all Impossible. Traditional search algorithms can only present the same sorting results to all users, and cannot provide corresponding services for different users’ interests. The explosion of information reduces the utilization of information, a phenomenon called Information overload. Personalized recommendation, including personalized search, is considered to be one of the most effective tools to solve the problem of information overload at present.
The recommendation problem is basically to evaluate products that it has never seen before on behalf of the users. These products include books, movies, CDs, web pages, and even restaurants, music, paintings, etc. — a process from the known to the unknown.