Application of recommendation system: An empirical study of the mobile reading platform

Abstract

Mobile reading on 'smart' terminals (like smartphones and tablet computers)is an increasing popular subject, and the recommendations of e-books for users also begin to attract more attentions. In this paper, we mainly demonstrate the performance of the personalized recommendation on the mobile reading platform, based on the analysis of the reading records on mobile phones. The analysis results of the feedback of users for the recommendations show that the personalized recommendation based on the mass diffusion algorithm is much better than the algorithm of the mobile company used before. In particular, both the number of the motivated page views and the motivated users have a dramatically increase. All these results indicate that the mass diffusion algorithm has an outstanding performance on the mobile reading recommendation, which can help users quickly find the books they are interested in. Meanwhile, it help the company enlarge the customer volume and improve the customer experience. © 2012 Springer-Verlag.