Publications
App Store Mining and Analysis (MSR2012)
In this project, we introduced the new idea of app store mining, a form of 'black box software repository mining'. We illustrated this idea with results from mining the 32,108 non-zero priced apps from the Blackberry app store. Our results show that there is a strong correlation between customer rating and the number of app downloads, though perhaps surprisingly, there is no correlation between price and downloads, nor between price and rating. These correlation findings carry over to (and are sometimes enhanced within) the space of data mined app features, thereby giving valuable information to the app developer about the relationship between apps' technical, customer and business aspects.
Results
Price vs Rating (sensitive distribution)
Price vs Rank of Downloads (sensitive distribution)
Rating vs Rank of Downloads (sensitive distribution)
Rating vs Rank of Downloads (insensitive distribution)
