Intruder Alert: Dimension Reduction and Density-Based Clustering for a Cybersecurity Application
Bibliographic record
Abstract
This thesis examines the use of Principal Component Analysis, Robust Principal Component Analysis, and simple autoencoders for dimension reduction on a synthetic cybersecurity dataset.Each is tested as a precursor to Independent Component Analysis.Stable independent components are obtained by iterative randomized starts to FastICA and selecting the centroids of the hierarchically clustered components.A density-based clustering method is then applied to the results with the goal of isolating malicious observations from benign ones using greatest distance between centroids as a heuristic metric of success.The method is then applied to a real-world cybersecurity dataset from an industry partner.i No project of this scale happens in a vacuum.To start with, thank you to Dr. Shirley Mills for taking a chance on me.Your kindness has been invaluable and your expertise irreplaceable.Thank you for supervising me and for introducing me to data mining.Thank you to my committee members, Drs.Dave Campbell and Aaron Smith, for your encouragement, your gentle questions, and your feedback on both the written portion and the defence.It was likely the nicest reviewer feedback anyone in academia could expect to see, and I will cherish that (and likely never publish again so that I can go out on a high note
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 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".