From PSAL

Privacy, Security and Automation Lab

Welcome to PSAL, Drexel University's Privacy, Security, and Automation Laboratory. We research on topics at the intersection between artificial intelligence, privacy and security, and human-computer interaction.

Check out the navigation sidebar for some basic information about who we are and what we are working on.

PI : Rachel Greenstadt (greenie at cs dot drexel dot edu)

JStylo / Anonymouth / Corpus Releases

Below are the links to the public alpha releases of JStylo and Anonymouth as well as separate downloads for each of the corpora we are releasing today. Note that if you download JStylo or Anonymouth they include these corpora and you do not need to re-download them.

Note: It is strongly recommended that you increase your java heap size when running JStylo and Anonymouth.

• Extended-Brennan-Greenstadt Adversarial Corpus (45 Authors, 6500 words per author minimum)
• Brennan-Greenstadt Adversarial Corpus (12 Authors, 5000 words per author minimum)

If you use this corpus in your research, please cite:

Michael Brennan, Sadia Afroz, and Rachel Greenstadt. 2012. Adversarial stylometry: Circumventing authorship recognition to preserve privacy and anonymity. ACM Trans. Inf. Syst. Secur. 15, 3, Article 12 (November 2012).

Michael Brennan and Rachel Greenstadt. Practical Attacks Against Authorship Recognition Techniques in Proceedings of the Twenty-First Conference on Innovative Applications of Artificial Intelligence (IAAI), Pasadena, California, July 2009.

JStylo / Anonymouth

If you use this software in your research, please cite:

Use Fewer Instances of the Letter "i": Toward Writing Style Anonymization. Andrew W.E. McDonald, Sadia Afroz, Aylin Caliskan, Ariel Stolerman and Rachel Greenstadt. Privacy Enhancing Technologies Symposium (PETS 2012), Vigo, Spain, July 2012

Recent News

• Paper accepted at Usenix Security 2015. Aylin Caliskan-Islam, Richard Harang, Andrew Liu, Arvind Narayanan, Clare Voss, Fabian Yamaguchi, and Rachel Greenstadt. De-anonymizing Programmers via Code Stylometry. 24th Usenix Security Symposium.
• Paper accepted at FC 2015. Vaibhav Garg, Sadia Afroz, Rebekah Overdorf and Rachel Greenstadt. Computer-Supported Cooperative Crime (Short Paper). Financial Cryptography and Data Security 2015.

• Paper accepted at WPES 2014. Aylin Caliskan-Islam, Jonathan Walsh, and Rachel Greenstadt. Privacy Detective: Detecting Private Information and Collective Privacy Behavior in a Large Social Network. Workshop on Privacy in the Electronic Society (WPES 2014).

• Paper accepted at ACM CCS 2014. A Critical Evaluation of Website Fingerprinting Attacks. Marc Juarez, Sadia Afroz, Gunes Acar, Claudia Diaz, Rachel Greenstadt.

• Paper accepted at HotPets 2014. Blogs and Twitter Feeds: A Stylometric Environmental Impact Study. Rebekah Overdorf, Travis Dutko, Rachel Greenstadt.

• Paper accepted at IEEE Symposium on Security & Privacy (Oakland 2014). Doppelgänger Finder: Taking Stylometry To The Underground code.

Sadia Afroz, Aylin Caliskan-Islam, Ariel Stolerman, Rachel Greenstadt, Damon McCoy.

• Two papers got accepted at IFIP WG11.9 2014:

1. Active linguistic authentication revisited: Real-time stylometric evaluation towards multi-modal decision fusion. Stolerman, A., Fridman, A., Greenstadt, R., Brennan, P. and Juola, P.

2. Classify, but Verify: Breaking the Closed-World Assumption in Stylometric Authorship Attribution. Stolerman, A., Overdorf, R., Afroz, S. and Greenstadt, R.

• Paper accepted at eCrime summit.Honor Among Thieves: A Common's Analysis of Cybercrime Economics.

Sadia Afroz, Vaibhav Garg, Damon McCoy, Rachel Greenstadt. Blog.

• Anonymouth was featured on The New Republic and Slashdot.

• We won the PET Award 2013 for the paper Adversarial Stylometry: Circumventing Authorship Recognition to Preserve Privacy and Anonymity.

• Paper accepted at PETools: Workshop on Privacy Enhancing Tools. Anonymouth Revamped: Getting Closer to Stylometric Anonymity. Andrew W.E. McDonald, Jeffrey Ulman, Marc Barrowclift and Rachel Greenstadt.

• Paper accepted at the 3rd workshop on Socio-Technical Aspects in Security and Trust (STAST '13). How Privacy Flaws Affect Consumer Perception. Sadia Afroz, Aylin Caliskan Islam, Jordan Santell, Aaron Chapin, Rachel Greenstadt.

• Paper accepted at the Ninth Annual IFIP WG 11.9 International Conference on Digital Forensics. A Dataset for Active Linguistic Authentication. Juola, P., Neocker, Jr. J., Stolerman, A., Ryan, M., Brennan, P. and Greenstadt, R. Ninth Annual IFIP WG 11.9 International Conference on Digital Forensics.

• Sadia and Aylin presented stylometric analysis of Online Underground Markets at 29c3.

This work has been featured in

1. Linguistics identifies anonymous users,

2. Stylometric analysis to track anonymous users in the underground

3. on Slashdot news, on Hacker news

4. Why hackers should be afraid of how they write

5. On Schneier on Security blog

• Congratulations to Dr. Brennan, PSAL's first PhD graduate.

• Paper accepted in the Fifth ACM workshop on Security and artificial intelligence (AISec '12). Robust Detection of Comment Spam Using Entropy Rate. Alex Kantchelian, Justin Ma, Ling Huang, Sadia Afroz, Anthony D. Joseph, J. D. Tygar.

• Paper accepted in the Sixth IEEE International Conference on Semantic Computing (ICSC 2012) "Translate once, translate twice, translate thrice and attribute: Identifying authors and machine translation tools in translated text" Aylin Caliskan and Rachel Greenstadt. Sixth IEEE International Conference on Semantic Computing (ICSC 2012).

• Michael Brennan won the best presentation award at HotPETS 2012 for "Perspectives on Academic Impact from Inside the Federal Trade Commission."

• "Use Fewer Instances of the Letter "i": Toward Writing Style Anonymization." won the best student paper award at PETS 2012.

• Congratulations to Michael Brennan for receiving the 2012 Jay Modi Award in recognition of academic excellence and the potential to become a leader in the field.

• Paper got accepted to PETS 2012. "Use Fewer Instances of the Letter "i": Toward Writing Style Anonymization." Andrew W.E. McDonald, Sadia Afroz, Aylin Caliskan, Ariel Stolerman and Rachel Greenstadt. The 12th Privacy Enhancing Technologies Symposium.

• Our paper got accepted to Oakland 2012. "Detecting Hoaxes, Frauds, and Deception in Writing Style Online." Afroz/Brennan/Greenstadt. IEEE Symposium on Security and Privacy '12.

• Our work was featured in NYTimes Bits blog, Boing Boing, and Der Spiegel (in German).

• Michael Brennan and Sadia Afroz presented at the 28th Chaos Communication Congress:video, slides.

• We will again be accepting participants for our Authorship Recognition Study soon.

• Sadia Afroz has a paper with Knarig Arabshian, Peter Danielsen at AAAI Spring Symposia 2012 on LexOnt: A Semi-automatic Ontology Creation Tool for Programmable Web.

• Sadia Afroz and Rachel Greenstadt. 2011. PhishZoo: Detecting Phishing Websites by Looking at Them. In Proceedings of the 2011 IEEE Fifth International Conference on Semantic Computing(ICSC '11). IEEE Computer Society, Washington, DC, USA, 368-375.

• PSAL and the Drexel Computer Science Department are hosting a Random Hacks of Kindness event on June 4th and 5th, 2011 (plus a reception on June 3rd). You can find more details and register at the RHoK Philadelphia registration page. It's free! Also, check out this article on Technically Philly about the event.

• Michael Brennan gave a talk in April at Indiana University's Center for Applied Cyber Security entitled Adversarial Stylometry: Attacking Authorship Recognition to Preserve Privacy and Anonymity.

• We have released our first adversarial stylometry corpus. It contains 12 of the 15 authors who made up the corpus used in "Practical Attacks Against Authorship Recognition Techniques." If you use this corpus in your research, please cite:

Michael Brennan and Rachel Greenstadt. Practical Attacks Against Authorship Recognition Techniques in Proceedings of the Twenty-First Conference on Innovative Applications of Artificial Intelligence (IAAI), Pasadena, California, July 2009.

• Michael Brennan, Stacey Wrazien, and Rachel Greenstadt have a paper at WikiAI (at AAAI) on Learning to Extract Quality Discourse in Online Communities Full feature tables

• Andrew Feit, Lenrik Toval, Raffi Hovagimian, and Rachel Greenstadt have a paper at the AAAI 2010 Workshop on Bridging The Gap Between Task And Motion Planning (BTAMP) on A Travel-Time Optimizing Edge Weighting Scheme for Dynamic Re-planning

• Rachel Greenstadt will be participating in the 2010 Workshop on Security and Human Behavior (speaking on community filtering in online communities) June 28-29.

• Rachel Greenstadt will be speaking at Future Challenges of Network Security (FNCS) in Prague on June 23 on Mixed Initiative Security Agents.

• Rachel Greenstadt and Michael Brennan have a poster at AAMAS on the Thursday, May 13 session (B-39). Come ask us questions about "Using Machine Learning to Augment Collaborative Filtering of Community Discussions"

• Rachel Greenstadt is giving a seminar talk on Privacy & Stylometry: Exploring the Limitations and Potential of Automated Authorship Recognition at the University of Waterloo at 13:30 pm on May 7.

• Michael Brennan passed his candidacy exam on "Current Trends and New Directions in Stylometry." He is now PSAL's first Ph.D. candidate.

• Rachel Greenstadt will chair the 3rd Workshop on Artificial Intelligence and Security (AISec) at the 17th ACM Computer and Communications Security Conference (CCS 2010) taking place October 4-8 in Chicago.

• Rachel Greenstadt is offering a new graduate level course on security. CS 645 will be offered Wednesday evenings and online.

• Michael Brennan spoke on Privacy and Stylometry: Practical Attacks on Authorship Recognition Techniques at the 26th Chaos Communication Congress in Berlin. slides mp4

• We presented a paper on Mixed Initiative Security Agents at the 2nd Workshop on Artificial Intelligence (AISec) at ACM CCS on November 9, 2009.

• Sadia Afroz was selected to present her anti-phishing research in the ACM Student Research Competition at the Grace Hopper conference. tech report.

• Paul Snyder, Rachel Greenstadt, and Giuseppe Valetto presented Myconet: A Fungi-inspired Model for Superpeer-based Peer-to-Peer Overlay Topologies at Self-Adaptive and Self-Organizing Systems (SASO 2009).

• Michael Brennan gave presentations on authorship recognition and privacy at HotPETs and Hacking at Random.

• Michael Brennan and Rachel Greenstadt's IAAI paper on foiling authorship recognition techniques [pdf] was featured in the New Scientist.

• Rachel Greenstadt and Bill Regli received an NSF grant to build a research exchange program with the Agent Technology Center at Czech Technical University (CTU) in Prague.