Discerning cyber threatening incidents from ordinary events using sentiment analysis and logistic regression
Bibliographic record
Abstract
Many organizations allow incident reports from the general public. Some of these reports may contain information about threatening incidents, while others may describe ordinary events. Incident classification is the process of distinguishing between incidents and events. We describe an automated incident classification system, which uses logistic regression and sentiment analysis to estimate the likelihood that an event is an incident using its textual description. We trained and validated two different models on one dataset and used a different dataset for testing purposes. The model that performed better utilized sentiment analysis at the sentence level as well as at the level of individual verbs, nouns, and adjectives. It achieved 99% accuracy on the validation set and 100% accuracy on the test set over 50% baseline. Overall, we found that using sentiment score increased the model's accuracy, precision, and recall by at least 10% especially when it is applied on several levels of the text. The difference between our approach and the typical human approach is that in our approach we train the system to recognize incidents before any incident actually takes place and our system can recognize incidents even if their descriptions do not include keywords the system previously encountered.
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 | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 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 | 0.001 |
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".