Upload dataset card
Browse files
README.md
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
task_categories:
|
| 5 |
+
- text-classification
|
| 6 |
+
tags:
|
| 7 |
+
- cybersecurity
|
| 8 |
+
- incident-classification
|
| 9 |
+
- technical-severity
|
| 10 |
+
- nis2
|
| 11 |
+
- cyber-blueprint
|
| 12 |
+
size_categories:
|
| 13 |
+
- 1K<n<10K
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# CyberScale Technical Severity Training Data
|
| 17 |
+
|
| 18 |
+
Training dataset for the CyberScale incident technical severity classifier (Phase 3 T-model). Contains 8,000 parametric incident scenarios with deterministic T1-T4 labels.
|
| 19 |
+
|
| 20 |
+
## Dataset Description
|
| 21 |
+
|
| 22 |
+
- **Source:** Parametrically generated from structured field combinations (50 templates x paraphrase variants)
|
| 23 |
+
- **Generation:** `training/scripts/generate_incidents.py`
|
| 24 |
+
- **Labels:** 4-class (T1-T4)
|
| 25 |
+
- **Balance:** 2,000 per class after balancing
|
| 26 |
+
|
| 27 |
+
## Schema
|
| 28 |
+
|
| 29 |
+
| Column | Type | Description |
|
| 30 |
+
|--------|------|-------------|
|
| 31 |
+
| `text` | string | Formatted input: `<description> [SEP] disruption: <level> entities: <N> sectors: <N> cascading: <level> data_compromise: <level>` |
|
| 32 |
+
| `label` | string | Technical severity level (T1/T2/T3/T4) |
|
| 33 |
+
|
| 34 |
+
## Class Distribution
|
| 35 |
+
|
| 36 |
+
| Label | Count |
|
| 37 |
+
|-------|-------|
|
| 38 |
+
| T1 | 2,000 |
|
| 39 |
+
| T2 | 2,000 |
|
| 40 |
+
| T3 | 2,000 |
|
| 41 |
+
| T4 | 2,000 |
|
| 42 |
+
|
| 43 |
+
## Citation
|
| 44 |
+
|
| 45 |
+
Part of the CyberScale project — multi-phase cyber severity assessment MCP server.
|