Cognitive vector quantization for malware detection
Bibliographic record
Abstract
In today’s world, detection of malware is a prevalent challenge due to the evolving nature of malware designs and techniques. Many machine learning algorithms, such as ANNs, use supervised learning, which relies on a labeled dataset. In real-time systems training sets are unlabeled and obtained in real-time. In this case, the use of unsupervised machine learning algorithms may be used. However, a problem with them is the generated clusters are not identified as either malware or benign. Much work has been done on incorporating cognition into ANNs to improve their performance. This thesis explores using a Vector Quantization Artificial Neural Network (VQ-ANN) to classify a malware dataset using unsupervised learning. Due to the unsupervised nature of the Vector Quantization Artificial Neural Network, the basic algorithm will not know the classification of examples during the training process and will attempt to sort the dataset based on similarities. This thesis uses a novel method to identify the clusters as either malware or benign by using elements of cognition in the form of the Variance Fractal Dimension to label the clusters formed in a VQ-ANN. As compared with currently used clustering methods (U-Matrix), our algorithm consistency produced a higher accuracy, with an average accuracy of 98.1%.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".