Skip to content

Latest commit

 

History

History
26 lines (14 loc) · 1.3 KB

README.md

File metadata and controls

26 lines (14 loc) · 1.3 KB

Minning iOS and Android App-pairs

A Large Scale Empirical Study on iOS and Android Mobile App-pairs.

Published as: Same App, Different App Stores: A Comparative Study, MobileSoft 2017. PDF: http://salt.ece.ubc.ca/publications/docs/mobilesoft17.pdf

The repo is split into two directories, where the "Data" directory contains a set of 80,169 app-pairs along with their attributes such as price, rating, stars etc.. The Data directory contains a txt file which has a dropbox link to a set of 2003 app-pairs along with their user reviews (We couldn't upload that dataset here due to github's file size restrictions). In addition to the reviews, for each app-pair, we include the results of our automated review classifiers and the success scores computed based on our proposed metric. Both of these datasets are JSON objects and can be imported directly to any mongoDB instance by using the following command: "mongorestore --collection insertCollectionNameHere --db inserDBNameHere pathToBsonFile"

The other directory "scripts" contains all the scripts we used to generate the data and proccess it.

We hope that by open sourcing our work we encourage others to conduct more research on mobile app-pairs.

If you have any questions or comments please don't hesitate to contact me at mohamed.ali92ubc@gmail.com

Thanks!