Using code metric histograms and genetic algorithms to perform author identification for software forensics

ABSTRACT

We have developed a technique to characterize software developers- styles using a set of source code metrics. This style fingerprint can be used to identify the likely author of a piece of code from a pool of candidates. Author identification has applications in criminal justice, corporate litigation, and plagiarism detection. Furthermore, we can identify candidate developers who share similar styles, making our technique useful for software maintenance as well. Our method involves measuring the differences in histogram distributions for code metrics.Identifying a combination of metrics that is effective in distinguishing developer styles is key to the utility of the technique. Our case study involves 18 metrics, and the time involved in exhaustive searching of the problem space prevented us from adding additional metrics. Using a genetic algorithm to perform the search, we were able to find good metric combinations in hours as opposed to weeks. The genetic algorithm has enabled us to begin adding new metrics to our catalog of available metrics. This paper documents the results of our experiments in author identification for software forensics and outlines future directions of research to improve the utility of our method.

REFERENCES

• 1: http://www.sourceforge.net/.
• 2: Stefan Berchtold , Christian Böhm , Daniel A. Keim , Hans-Peter Kriegel, A cost model for nearest neighbor search in high-dimensional data space, Proceedings of the sixteenth ACM SIGACT-SIGMOD-SIGART symposium on Principles of database systems, p.78-86, May 11-15, 1997, Tucson, Arizona, United States [doi>10.1145/263661.263671]
• 3: R. Bouckaert. Bayesian network classifiers in weka. Technical Report 14/2004, The University of Waikato, Department of Computer Science, Hamilton, New Zealand, 2004.
• 4: Gülsen Demiröz , H. Altay Güvenir, Classification by Voting Feature Intervals, Proceedings of the 9th European Conference on Machine Learning, p.85-92, April 23-25, 1997
• 5: Haibiao Ding , Mansur H. Samadzadeh, Extraction of Java program fingerprints for software authorship identification, Journal of Systems and Software, v.72 n.1, p.49-57, June 2004 [doi>10.1016/S0164-1212(03)00049-9]
• 6: David E. Goldberg, Genetic Algorithms in Search, Optimization and Machine Learning, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1989
• 7: A. Gray , P. Sallis , S. MacDonell, IDENTIFIED: A Dictionary-Based System for Extracting Source Code Metrics for Software Forensics, Proceedings of the 1998 International Conference on Software Engineering: Education & Practice, p.252, January 26-29, 1998
• 8: J. Lenahan. Pygene Open Source Evolutionary Computation Tool. SIGEVOlution Newsletter, 1(2):27, 2006.
• 9: S. Macdonell, A. Gray, G. MacLennan, and P. Sallis. Software forensics for discriminating between program authors usingcase--based reasoning, feedforward neural networks and multiplediscriminant analysis. Neural Information Processing, 1999. Proceedings. ICONIP'99. 6th International Conference on, 1, 1999.
• 10: P. W. Oman , C. R. Cook, Programming style authorship analysis, Proceedings of the 17th conference on ACM Annual Computer Science Conference, p.320-326, February 21-23, 1989, Louisville, Kentucky [doi>10.1145/75427.75469]
• 11: J. Ross Quinlan, C4.5: programs for machine learning, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1993
• 12: P. Sallis. Contemporary Computing Methods for the Authorship Characterisation Problem in Computational Linguistics. New Zealand Journal of Computing, 5(1):85--95, 1994.
• 13: P. Sallis, S. MacDonell, G. MacLennan, A. Gray, and R. Kilgour. Identified: Software authorship analysis with case--based reasoning. Proc. Addendum Session Int. Conf. Neural Info. Processing and Intelligent Info. Systems, pages 53--56, 1997.
• 14: E. Spafford and S. Weeber. Software forensics: Can we track code to its authors. Technical Report CSD-TR 92-010, Purdue University, Dept. of Computer Sciences, 1992.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.6 Learning

General Terms:
Experimentation, Languages, Measurement

Keywords:
author identification, genetic algorithm applications, pygene, software forensics, software metrics