Design And Implementation Of General Recommendation Service Framework Supporting Algorithm Plugin

With the application and development of Internet technology and the explosive development of modern e-commerce, information on the Web is growing exponentially. Today, the quantity of goods and services that users can access over the Internet reached an unprecedented scale. While vast amounts of information is presented, but only a small part is interested to the user. Nowadays, people rely on recommender system to solve the problem of information overload.Personalized recommender systems are widely applied to various fields. However, conventional systems are often developed for a particular application field, the way it provided is often closed and inflexible. In this paper, a framework for recommender system is proposed, which support various algorithms by plugins and encapsulates the collection and analysis of data, and high performance recommender interface. With a simple interface application developers can get reliable and efficient recommendation capabilities, allowing them focusing on application development, reduce development time.In this paper, the principles and concepts recommender system is introduced. The general structure of the recommendation system is presented, and the requirements analysis of the general framework for recommender system(GFRS) is made. Then it comes to the details of the design and implementation of GFRS. In the end, the recommender module development process of a sample site is used to illustrate the effectiveness and availability of the framework.