Insider Threat Detection using Profiling and Cyber-persona Identification
Bibliographic record
Abstract
Nowadays, insider threats represent a significant concern for government and business organizations alike. Over the last couple of years, the number of insider threat incidents increased by 47%, while the associated cost increased by 31%. In 2019, Desjardins, a Canadian bank, was a victim of a data breach caused by a malicious insider who exfiltrated confidential data of 4.2 million clients. During the same year, Capital One was also a victim of a data breach caused by an insider who stole the data of approximately 140 thousand credit cards. Thus, there is a pressing need for highly-effective and fully-automatic insider threat detection techniques to counter these rapidly increasing threats. Also, after detecting an insider threat security event, it is essential to get the full details on the entities causing it and to gain relevant insights into how to mitigate and prevent such events in the \nfuture. In this thesis, we propose an elaborated insider threat detection system leveraging user profiling and cyber-persona identification. We design and implement the system as a framework that employs a combination of supervised and unsupervised machine learning and deep learning techniques, which allow modelling the normal behaviour of the insiders passively by analyzing their network traffic. We can deploy the framework as part of online traffic monitoring solutions for insider profiling and cyber-persona identification as well as for detecting anomalous network behaviours. The different models employed are assessed \nusing specific metrics such as Accuracy, F1 score, Recall and Precision. The conducted experimental evaluation indicates that the proposed framework is efficient, scalable, and suitable for near-real-time deployment scenarios.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".