Mobile App Store Analytics


NII Shonan Meeting Seminar 070


Today, software engineering research focuses on traditional software?systems like the Firefox web browser or Microsoft Windows, which take?years to develop and teams of designers, developers and testers.?Software engineering is rapidly changing though. Emerging domains,?such as mobile devices, are growing rapidly and depend heavily on new?software operating systems like Android and the applications that they?run, commonly referred to as apps. Over the past few years, we have?seen a boom in the popularity of mobile devices and mobile apps which?run on these devices. Recent market studies predict that the global?mobile app economy is expected to be worth $143 billion by 2016. Thus?there exists considerable motivation for the research community to?solve the challenges faced by the mobile app developers.

However, unlike traditional software, the distribution mechanism for?mobile apps are very different ? they are released through app markets?(e.g., Google Play and Appleā€™s App store). The key differentiating?factor in an app store is that it is a democratic platform, i.e., both?large companies with established products Adobe Reader from Adobe, and?Timberman from Digital Melody (a company with 5 employees), can?release their apps through the same mechanism for the users to?download and install. The data that these mobile app markets contain?can be used by software engineering researchers to compile new?empirical results that can help mobile app developers.

Additionally, the app markets allow users to post reviews of the apps.?This is very different from traditional software. Mobile app?developers get continuous feedback from users that can be leveraged to?help them. For example, prior work leveraged user reviews to extract?user-faced issues, and new requirements. However, today the review?system for mobile apps is identical to that of books sold on an?e-commerce website such as Amazon. While books are products too, they?are very different from mobile apps in that books are not updated?every few weeks like most mobile apps. Therefore in the case of books,?the ratings and reviews collected for a book is all with respect to?one version of a book, while the ratings and reviews collected for an?app are about all the versions of an app. Hence the question arises?whether the review systems of books is the best system for mobile apps?or not?

Finally, the app stores provide a central location for all the apps,?making it easy for researchers to mine the store for meta-data of the?apps and the apps themselves. Using the data from the app stores,?several companies like App Annie, and Distimo have even built?successful businesses selling intelligence gained from observing the?evolution of several hundred thousand apps in the app stores.

However, there is no central venue to bring all the cross-disciplinary?researchers together. There is therefore, a dire need for a community?to be built around the line of research with respect to mobile app?store analytics. The proposed seminar would focus on research where?the Mobile App Stores, and the data that they have are mined for?insights into mobile app development. We intend to bring researchers?in multiple disciplines from around the world in one place to discuss?the future directions in the area of mobile app store analytics.


Category: Uncategorized


Comments are closed.