Predicting Vulnerable Software Components via Text Mining
Riccardo Scandariato, James Walden, Aram Hovsepyan et al. · IEEE Transactions on Software Engineering · 2014
This paper presents an approach based on machine learning to predict which components of a software application contain security vulnerabilities. The approach is based on text mining the source code of the components. Namely, each component is characterized as a series of terms contained in its source code, with the associated frequencies. These features are used to forecast whether each component is likely to contain vulnerabilities. In an exploratory validation with 20 Android applications, we discovered that a dependable prediction model can be built. Such model could be useful to prioritize the validation activities, e.g., to identify the components needing special scrutiny.
Cited by 337 works. doi:10.1109/tse.2014.2340398 Open-access PDF
Authors
References
- Software Engineering Economics
- A few useful things to know about machine learning
- The Lack of A Priori Distinctions Between Learning Algorithms
- On biases in estimating multi-valued attributes
- Evaluating Complexity, Code Churn, and Developer Activity Metrics as Indicators of Software Vulnerabilities
- Classifying Software Changes: Clean or Buggy?
- Cross-project defect prediction: a large scale experiment on data vs. domain vs. process
- Predicting vulnerable software components
- A systematic review of software fault prediction studies
- On the value of static analysis for fault detection in software