Detecting Concept Drift in Evolving Malware Families Using Rule-Based Classifier Representations
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913059242508288 |
|---|---|
| author | Kalný, Tomáš Jureček, Martin Stamp, Mark |
| author_facet | Kalný, Tomáš Jureček, Martin Stamp, Mark |
| contents | This work proposes a structural approach to concept drift detection in malware classification using decision tree rulesets. Classifiers are trained across temporal windows on the EMBER2024 dataset, and drift is quantified by comparing extracted rule representations using feature importance, prediction agreement, activation stability, and coverage metrics. These metrics are correlated with both accuracy degradation and data distribution shift as complementary drift indicators. The approach is evaluated across six malware families using fixed-interval and clustering-based windowing in family-vs-benign and family-vs-family settings, and compared against RIPPER and Transcendent baselines. Results show that fixed two-month windowing with feature-level Pearson correlation is the most reliable configuration, being the only one where all family pairs produce positive drift-accuracy correlations. The methods are complementary - no single approach dominates across all pairs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_22629 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Detecting Concept Drift in Evolving Malware Families Using Rule-Based Classifier Representations Kalný, Tomáš Jureček, Martin Stamp, Mark Cryptography and Security Machine Learning This work proposes a structural approach to concept drift detection in malware classification using decision tree rulesets. Classifiers are trained across temporal windows on the EMBER2024 dataset, and drift is quantified by comparing extracted rule representations using feature importance, prediction agreement, activation stability, and coverage metrics. These metrics are correlated with both accuracy degradation and data distribution shift as complementary drift indicators. The approach is evaluated across six malware families using fixed-interval and clustering-based windowing in family-vs-benign and family-vs-family settings, and compared against RIPPER and Transcendent baselines. Results show that fixed two-month windowing with feature-level Pearson correlation is the most reliable configuration, being the only one where all family pairs produce positive drift-accuracy correlations. The methods are complementary - no single approach dominates across all pairs. |
| title | Detecting Concept Drift in Evolving Malware Families Using Rule-Based Classifier Representations |
| topic | Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2604.22629 |