Prediction of the Number of Downloads Of Smartphone Applications Based on the Markov Chain Approach

10.6702/ijbi.2015.10.1.1

Satoshi Mizusawa；Ushio Sumita；Maiko Takano

Smartphone applications ； number of devices ； probability distributions ； Markov chain approach

International Journal of Business and Information

10卷1期（2015 / 03 / 01）

1 - 23

繁體中文

Although the smartphone application market has been growing with amazing speed, the study of the market is still in its infancy. Perhaps the major reason for this fact is the difficulty of obtaining real data concerning how such applications are downloaded and removed by users along the time axis. Recently, Fuller, Inc., in Japan developed a software package called ＂Uncle Smartphone.＂ The corporation operates a software development company; develops, manufactures, and sells mobile phone applications; and also provides Internet media services. ＂Uncle Smartphone＂ enables the company to acquire information about which smartphone applications are downloaded to and removed from the devices that are registered with the customer's agreement. Using this data set, Sumita Research Laboratory at the University of Tsukuba has been conducting a joint research project with Fuller, Inc., to explore the smartphone application market in Japan. As part of the accomplishments that this joint project has achieved so far, this paper develops an algorithm for estimating the probability distribution of the number of downloads of a smartphone application based on the Markov chain approach. This kind of information is quite useful for smartphone application developers, since, for example, it becomes possible to understand the life cycle of their applications. Numerical results are presented based on real data, demonstrating the validity of the proposed algorithm.

1. Yano Research Institute. 2012. Study of Smartphone Application Market in Japan, downloadable from website: http://www.yano.co.jp/press/pdf/919.pdf
2. Uslu, A. and T. Cam. 2014. Analysis of Brand Loyalty with Markov Chains, downloadable from website: http://www.opf.slu.cz/vvr/akce/turecko/pdf/Uslu.pdf
3. MM Research Institute. 2012. Estimation of the Smartphone Application Market in Japan, downloadable from website http://www.m2ri.jp/newsreleases/main.php?id=010120131009500
4. Svanberg, J. 2013. The Mobile Application Market, BERG INSGHT, downloadable from website http://www.berginsight.com/ReportPDF/ProductSheet/bi-app2-ps.pdf
5. Hideyuki T.,Ushio, S.(2011).Credit risk model with contagious default dependencies affected by macro-economic condition.European Journal of Operational Research,214,365-379.
6. Jun, Y.(2013).Graduate School of Systems and Information Engineering, University of Tsukuba.
7. Keilson, J.(1979).Markov Chain Models-Rarity and Exponentiality.New York:Springer-Verlag.
8. Khajvand, M.,Mohammad, J.T.(2011).Estimating customer future value of different customer segments based on adapted RFM model in retail banking context.Procedia Computer Science,2011(3),1327-1332.
9. Masaji, K.、Minoru, E.(2012)。Estimation of influencers and application usage based on historical data of application usage。NTT DOCOMO Technical Journal，20(2)
10. Ushio, S.,Huang, J.P.(2009).Dynamic analysis of a unified multivariate counting process and its asymptotic behavior.International Journal of Mathematics and Mathematical Sciences
11. Ushio, S.,Takahashi, K.,Sugaya, K.,Saito, K.,Koike, Y.(2007).Analysis of brand characteristics of music artists.The Operations Research Society of Japan,69-78.
12. Ushio, S.,Yoshii, J.(2012).Impact of mobile access to the Internet on sales completion time in e-commerce.Journal of Modern Accounting and Auditing,8(4),503-528.
13. Ushio, S.,Yoshii, J.(2010).Enhancement of e-commerce via mobile accesses to the Internet.Electric Commerce Research and Applications
14. Zhu, Jianhan,Hong, Jun,Hughes, John G.(2002).Using Markov Chains for Link Prediction in Adaptive Web Sites.Soft-Ware 2002