Taint Reverse Propagation for Analysis of Privacy Leak

摘要

Privacy leak is becoming a serious problem for every computer user. It will be a threat to user accounts, passwords and private documents. Privacy leak detection has been an important research point in security research community. Current privacy leak techniques mainly focus on static analysis and dynamic taint tracking. Static analysis often produces too many false positives. Dynamic taint tracking will degrade the performance of program execution. In the paper, we propose a method which uses taint reverse propagation to analyze the data streams in programs. Experiments show that we can achieve more efficient and accurate analysis for the analysts to judge the programs' behaviors. To the best of our knowledge, we are the first team to use the reverse propagation in privacy leak.