Design And Implementation Of Collaborative Filtering Engine In Mobile Reading Real-time Recommendation System

With the popularity of smart phones and the development of mobile internet technology, more and more users choose to read books on their mobile phones. Facing the tremendous amount of e-books online, how to help users quickly and efficiently find the books what they really want according to their interests and enhance the user’s reading experience has gradually become an important demand in mobile reading industry. It is in this context, combined with the previously learned off-line recommendation system of high delay, slow response, complex operation.and maintenance and other shortcomings, we propose a real-time recommendation project of mobile reading.This thesis is based on the BI mobile platform. According to the user’s behavior of ordering, reading and browsing, we designed and implemented the recommendation system on the basis of open source data framework of Hadoop and Storm with the collaborative filtering algorithm. The system has the following characteristics, first of all, it has the characteritics of real-time, which can provide users with relevant recommendations fast and efficiently according to the user’s real-time behavior,greatly enhance the user’s reading experience; second, the system has high availability, scalability characteristics, can be a good response to the increase in the number of users with high concurrent requests, which improve the stability of the system; last but not the least,the system has the characteristics of versatility, which is flexible of to a variety of recommendation algorithms and lays the foundation of adding new recommendation algorithms to the system afterwards.After the completion of the design and implementation of the system, we conducted a functional test of the system at first, then we tested the performance of the components in the system, at last, we use the A/B test method to verify the common indicators of recommendation system, like accuracy and diversity, which to gurantee the reliability and validity of the recommendation system. At present, the recommendation system described in this thesis has been successfully running in the background of migu reading app, and has brought a good income for the mobile platform.