ABSTRACT
The explosive growth of the mobile application market has made it a significant challenge for the users to find interesting applications in crowded App Stores. To alleviate this problem, existing industry solutions often use the users' application download history and possibly their ratings to recommend applications that might interest them, much like Amazon's book recommendations. However, the user downloading an application is a weak indicator of whether the user likes that application, particularly if the application is free and the user just wants to try it out. Using application ratings, on the other hand, suffers from tedious manual input and potential data sparsity problems.
In this paper, we present the AppJoy system that makes personalized application recommendations by analyzing how the user actually uses her installed applications. Based on all participants' application usage records, AppJoy employs an item-based collaborative filtering algorithm for individualized recommendations. We discuss AppJoy's design and implementation, and the evaluation shows that it consumes little resource on the off-the-shelf Google Android phones. AppJoy has been available in the Android Market and used by more than 4600 users. The AppJoy's prediction algorithm provided reasonably accurate usage estimate of the recommended applications after they were installed. We also found AppJoy to be effective as the users interacted with recommended applications longer than other applications.
- AppBrain. http://www.appbrain.com/.Google Scholar
- AppsFire. http://www.appsfire.com/.Google Scholar
- Google Android Developers. http://developer.android.com/.Google Scholar
- J. Bennett and S. Lanning. The Netflix Prize. In The ACM SIGKDD Cup and Workshop, 2007.Google Scholar
- J. R. Bult and T. Wansbeek. Optimal selection for direct mail. Marketing Science, 14(4), 1995.Google Scholar
- K. Church and B. Smyth. Understanding mobile information needs. In Proceedings of MobileHCI, 2008. Google ScholarDigital Library
- H. Falaki, R. Mahajan, S. Kandula, D. Lymberopoulos, R. Govindan, and D. Estrin. Diversity in smartphone usage. In Proceedings of MobiSys, 2010. Google ScholarDigital Library
- G. Holmberg and M. Torrens. Musicstrands: A platform for discovering and exploring music. University of Michigan Library, 2005.Google Scholar
- J. Layton. How Pandora Radio works. 2006.Google Scholar
- D. Lemire and A. Maclachlan. Slope one predictors for online rating-based collaborative filtering. In Proceedings of SIAM on Data Mining, 2005.Google ScholarCross Ref
- Why the mobile Web is disappointing end-users. Equation Research Report, Oct. 2009.Google Scholar
- S. Perez. Mobile app marketplace. ReadWriteWeb.com, Mar. 2010.Google Scholar
- B. Sarwar, G. Karypis, J. Konstan, and J. Riedl. Item-based collaborative filtering recommendation algorithms. In Proceedings of WWW, 2001. Google ScholarDigital Library
- J. B. Schafer, J. Konstan, and J. Riedi. Recommender systems in E-commerce. In Proceedings of the 1st ACM Conference on Electronic Commerce (ACM-EC), 1999. Google ScholarDigital Library
- C. Shepard, C. Tossel, A. Rahmati, L. Zhong, and P. Kortum. LiveLab: Measuring wireless networks and smartphone users in the field. In Proceedings of HotMetrics, 2010. Google ScholarDigital Library
- T. Sohn, K. A. Li, W. G. Griswold, and J. D. Hollan. A diary study of mobile information needs. In Proceedings of CHI, 2008. Google ScholarDigital Library
- A. Toscher, M. Jahrer, and R. M. Bell. The BigChaos solution to the Netflix Grand Prize, 2009.Google Scholar
- W. Woerndl, C. Schueller, and R. Wojtech. A hybrid recommender system for context-aware recommendations of mobile applications. In Proceedings of the IEEE ICDE, 2007. Google ScholarDigital Library
Index Terms
- AppJoy: personalized mobile application discovery
Recommendations
Personalized Recommendation Algorithm Using User Demography Information
WKDD '09: Proceedings of the 2009 Second International Workshop on Knowledge Discovery and Data MiningPersonalized recommendation systems are web-based systems that aim at predicting a user’s interest on available products and services by relying on previously rated items and dealing with the problem of information and product overload. User demography ...
Personalized rough-set-based recommendation by integrating multiple contents and collaborative information
In recent years, explosively-growing information makes the users confused in making decisions among various kinds of products such as music, movies, books, etc. As a result, it is a challenging issue to help the user identify what she/he prefers. To ...
Representation learning with collaborative autoencoder for personalized recommendation
AbstractIn the past decades, recommendation systems have provided lots of valuable personalized suggestions for the users to address the problem of information over-loaded. Collaborative Filtering (CF) is one of the most commonly applied and successful ...
Highlights- Two different autoencoders are used to capture characteristics for users and items.
- Manifold regularization is integrated into autoencoder for user’s features learning.
- The comprehensive experiments evaluate the effectiveness of ...
Comments