TWOS: A Dataset of Malicious Insider Threat Behavior Based on a Gamified Competition
The DOI matches its Crossref record; 25 of the 25 references it cites were found in a registry or an index.
Abstract
In this paper we present the design and outcome of a gamified competition that was devised in order to obtain a dataset containing realistic instances of insider threats. The competition simulated user interactions in/among competing companies, where two types of behaviors (normal and malicious) were incentivized. For the case of malicious behavior, we designed sessions for two types of insider threats (masqueraders and traitors). The game involved the participation of 6 teams consisting of 4 students who competed with each other for a period of 5 days, while their activities were monitored considering several heterogeneous sources (mouse, keyboard, process and file-system monitor, network traffic, emails and login/logout). In sum, we obtained 320 hours of active participation that included 18 hours of masquerader data and at least two instances of traitor data. Additionally to malicious behaviors, the students explored various defensive and offensive strategies, such as denial of service attacks and obfuscation techniques, in an effort to get ahead in the competition. The TWOS dataset is publicly accessible for further research purposes.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The TWOS dataset, generated through a gamified competition [85], provides authentic insider threat instances. This publicly available dataset includes data on masquerader and traitor behaviors, defensive and offensive strategies, making it a valuable resource for malicious threat detection research.”
From A Review of Recent Advances, Challenges, and Opportunities in Malicious Insider Threat Detection Using Machine Learning Methods (2024) · cites this paper as [85]
Citations
Cites
The dark triad of personality: Narcissism, Machiavellianism, and psychopathy
Paulhus, D. L., & Williams, K. M. (2002) · Journal of Research in Personality · doi:10.1016/s0092-6566(02)00505-6 · Cited by 5,265 · cited 1×
“We used the questionnaire inspired by dark triad theory [37], which contained 50 questions.”
MethodSabotage in the workplace: The role of organizational injustice
Ambrose, M. L., Seabright, M. A., & Schminke, M. (2002) · Organizational Behavior and Human Decision Processes · doi:10.1016/s0749-5978(02)00037-7 · Cited by 702 · cited 1×
“Taking inspiration from the previous works [5, 15, 17, 24, 29, 47], we asked the participants to fill up a psychological questionnaire.”
MethodComparing anomaly-detection algorithms for keystroke dynamics
Killourhy, K. S., & Maxion, R. A. (2009) · 2009 IEEE/IFIP International Conference on Dependable Systems & Networks · doi:10.1109/dsn.2009.5270346 · Cited by 566 · cited 1×
“As an effort to combat these threats, several works have proposed the analysis of user behavior using various features (i.e., file-system interaction [13, 14], biometric behavior by means of mouse [20] and keyboard usage patterns [26], among others).”
BackgroundComputer Intrusion: Detecting Masquerades
DuMouchel, W., Ju, W.-H., Karr, A. F., Schonlau, M., Theusan, M., & Vardi, Y. (2001) · Statistical Science · doi:10.1214/ss/998929476 · Cited by 378 · cited 3×
“• mouse activity [19], • keystrokes [33], • host activities [12, 16, 44] (such as system calls or file system interaction), • network traffic [16], • email activity [11, 16], • login and Logout [16, 22], • UNIX commands [39, 46].”
BackgroundBridging the Gap: A Pragmatic Approach to Generating Insider Threat Data
Glasser, J., & Lindauer, B. (2013) · 2013 IEEE Security and Privacy Workshops · doi:10.1109/spw.2013.37 · Cited by 344 · cited 1×
“CERT with other partners generated a collection of synthetic insider threat datasets [16] and described generation approach of the datasets in [21].”
BackgroundInternet Security Glossary, Version 2
Shirey, R. (2007) · doi:10.17487/rfc4949 · Cited by 300
Solomon, D. A., & Custer, H. (1992) · Medical Entomology and Zoology · Cited by 292
A Survey of Insider Attack Detection Research
Salem, M. B., Hershkop, S., & Stolfo, S. J. (2008) · Advances in Information Security · doi:10.1007/978-0-387-77322-3_5 · Cited by 280 · cited 8×
“From this statement, we have defined our attacker models by taking inspiration from ones already discussed in the literature: masquer-ader and traitor [41].”
BackgroundMethodInfluentialMasquerade detection using truncated command lines
Maxion, R. A., & Townsend, T. N. (2003) · Proceedings International Conference on Dependable Systems and Networks · doi:10.1109/dsn.2002.1028903 · Cited by 221 · cited 1×
“Maxion showed that the Schonlau dataset is not appropriate for the masquerader detection task [30].”
MethodAn Application of Machine Learning to Anomaly Detection
Lane, T. D. R. (1999) · Cited by 188
Cited by
Deep learning for insider threat detection: Review, challenges and opportunities
Yuan, S., & Wu, X. (2021) · Computers & Security · doi:10.1016/j.cose.2021.102221 · Cited by 271
Homoliak, I., Toffalini, F., Guarnizo, J., Elovici, Y., & Ochoa, M. (2020) · ACM Computing Surveys · doi:10.1145/3303771 · Cited by 255
“Furthermore, the authors presented applicable state-of-the-art features and demonstrated the potential use of the TWOS dataset in multiple areas of cyber-security that relate to the insider threat area, such as authorship verification and identification, continuous authentication, and sentiment analysis [Harilal et al…”
BackgroundInfluentialSelf-citationAlzaabi, F. R., & Mehmood, A. (2024) · IEEE Access · doi:10.1109/access.2024.3369906 · Cited by 204
“TheTWOS dataset, generated through a gamified competition [85], provides authentic insider threat instances.”
BackgroundAl-Mhiqani, M. N., Ahmad, R., Zainal Abidin, Z., Yassin, W., Hassan, A., Abdulkareem, K. H., Ali, N. S., & Yunos, Z. (2020) · Applied Sciences · doi:10.3390/app10155208 · Cited by 169
“, keystrokes, mouse, host monitor, network traffic, SMTP logs, and logon), with additional data from a psychological personality questionnaire [191].”
BackgroundMethodUser Authentication Based on Mouse Dynamics Using Deep Neural Networks: A Comprehensive Study
Chong, P., Elovici, Y., & Binder, A. (2019) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2930429 · Cited by 96
“We use the same datasets as the original work: the Balabit Mouse Dynamics Challenge [25] and the Wolf of SUTD (TWOS) [26] datasets.”
Method
Metadata
- Authors
- Harilal, A., Toffalini, F., Castellanos, J., Guarnizo, J., Homoliak, I., & Ochoa, M.
- Affiliations
- Singapore University of Technology and Design (SG)
- Venue
- Proceedings of the 2017 International Workshop on Managing Insider Security Threats
- Publisher
- ACM
- Published
- 2017-10-30
- Pages
- 45-56
- OpenAlex
- W2766640141
- Cited by
- 53
- References
- 25
- Citations by year
- 2026: 5 · 2025: 5 · 2024: 11 · 2023: 5 · 2022: 8 · 2021: 7 · 2020: 5 · 2019: 5 · 2018: 2
- FWCI
- 2.52
- Type
- conference paper
- Language
- en
- Open access
- closed
- Funders
- National Research Foundation
- ST Electronics
- Topics
- Physical Sciences › Computer Science › Information Systems › Information and Cyber Security
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Physical Sciences › Computer Science › Information Systems › User Authentication and Security Systems
- Keywords
- insider threats, masquerading, dataset collection, user behavior simulation, malicious user behavior, defense strategy, denial-of-service attacks, network traffic monitoring, file system monitoring
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: TWOS: A Dataset of Malicious Insider Threat Behavior Based on a Gamified Competition. https://app.cochat.ai/papers/10.1145/3139923.3139929
- The paper
- Harilal, A., Toffalini, F., Castellanos, J., Guarnizo, J., Homoliak, I., & Ochoa, M. (2017). TWOS: A Dataset of Malicious Insider Threat Behavior Based on a Gamified Competition. Proceedings of the 2017 International Workshop on Managing Insider Security Threats. https://doi.org/10.1145/3139923.3139929
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.