OASIS Project

OASIS project aims to prioritize static analyzer warnings for Android apps based on App user reviews. OASIS prioritize the warnings based on the observation:
Warnings and user reviews are intrinsically correlated

  • A warning should be ranked higher if its described issue can cause user-perceivable problems
  • Recover links between warnings and issues described in app user reviews to estimate the user-perceivable problems caused by a warning

Related publication

  1. Lili Wei, Yepang Liu, and Shing-Chi Cheung. OASIS: Prioritizing Static Analysis Warnings for Android Apps Based on App User Reviews. In Proceedings of the 11th joint meeting of the European Software Engineering Conference and the ACM SIGSOFT Symposium on the Foundations of Software Engineering (ESEC/FSE 2017), Paderborn, Germany, Sept 2017
    A preprint is available here.


Should you have any questions please contact: lweiae AT cse DOT ust DOT hk


Our Tool and Dataset
Please note that the tool is very large due to the libraries we used.

The current implementation of OASIS contains two parts: an Android Studio Plugin that augment the warnings and a jar file including a standalone prioritizer taking the outputs of the plugin as input. To run OASIS, you need to follow the following steps:

We also provide our dataset used in our experiments, including the original warnings exported from Android Studio, the reviews we crawled in JSON format, categorized reviews output by SURF in html format, intermediate outputs generated by the plugin and the positive warnings we identified.

Terms of Use

The data set and tool are released for only acedemic or personal use. We would very much appreciate if you accredit us when making use of the released materials.