README.md

---
license: cc-by-nc-4.0
task_categories:
- tabular-classification
- time-series-forecasting
tags:
- tsfile
- timeseries
- cybersecurity
- siem
- security-logs
- mitre-attack
- apt
- synthetic-data
- alert-triage
- soc-operations
- threat-detection
- splunk
- time-series
- format:tsfile
pretty_name: CYB010 — Synthetic Security Event Log Dataset (Sample) (TsFile)
size_categories:
- 10K<n<100K
configs:
- config_name: default
  data_files:
  - split: alert_records
    path: alert_records.tsfile
  - split: host_inventory
    path: host_inventory.tsfile
  - split: incident_summary
    path: incident_summary.tsfile
  - split: security_events
    path: security_events.tsfile
---

# CYB010 — Synthetic Security Event Log Dataset (Sample) (TsFile)

Apache TsFile version of [`xpertsystems/cyb010-sample`](https://huggingface.co/datasets/xpertsystems/cyb010-sample).

## Overview

**CYB010 — Synthetic Security Event Log Dataset (Sample)** — This is a **free preview** of the full **CYB010 — Synthetic Security Event Log Dataset** product. It contains roughly **~10% of the full dataset** at identical schema, MITRE ATT&CK technique coverage, and statistical fingerprint, so you can evaluate fit before licensing the full product.

- **Tables:** `alert_records`, `host_inventory`, `incident_summary`, `security_events`
- **Rows:** 27,958 across 4 TsFile table(s).
  - `alert_records`: 5,162 rows, 2024-01-02 20:47:00.595000 → 2025-03-23 01:53:19.867000 (UTC)
  - `host_inventory`: 400 rows — Time is synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
  - `incident_summary`: 500 rows, 2024-01-01 00:56:18.777000 → 2024-12-30 18:22:58.977000 (UTC)
  - `security_events`: 21,896 rows, 2024-01-01 00:56:29.283000 → 2025-04-10 03:36:47.436000 (UTC)

## Schema (TsFile structure)

### `alert_records.tsfile`

- **Time** (INT64, milliseconds) — the source `timestamp` timestamp as INT64 milliseconds.
- **TAG**: `host_id` — device dimension(s); query one device with `WHERE host_id=...`.
- **FIELD** (19): `alert_id` (string), `correlated_event_ids` (string), `alert_rule_id` (string), `alert_rule_name` (string), `user_id` (string), `threat_actor_id` (string), `severity_level` (string), `alert_category` (string), `mitre_technique_id` (string), `mitre_tactic` (string), `label_true_positive` (int64), `label_false_positive` (int64), `analyst_triage_priority` (string), `suppressed_flag` (int64), `suppression_reason` (string), `time_to_detect_seconds` (double), `siem_platform_type` (string), `correlated_chain_length` (int64), `incident_id` (string)

### `host_inventory.tsfile`

- **Time** (INT64, milliseconds) — synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
- **TAG**: `host_id` — device dimension(s); query one device with `WHERE host_id=...`.
- **FIELD** (14): `hostname` (string), `ip_address` (string), `os_type` (string), `os_version` (string), `host_role` (string), `network_segment` (string), `defender_posture_tier` (string), `edr_agent_installed` (int64), `patch_compliance_level` (double), `criticality_rating` (string), `cloud_provider` (string), `cloud_region` (string), `vulnerability_count_open` (int64), `siem_platform` (string)

### `incident_summary.tsfile`

- **Time** (INT64, milliseconds) — the source `campaign_start_timestamp` timestamp as INT64 milliseconds.
- **TAG**: `incident_id` — device dimension(s); query one device with `WHERE incident_id=...`.
- **FIELD** (22): `threat_actor_id` (string), `threat_actor_profile` (string), `campaign_end_timestamp` (string), `dwell_time_hours` (double), `attack_lifecycle_phase_reached` (string), `total_events_generated` (int64), `total_alerts_triggered` (int64), `true_positive_alerts` (int64), `false_positive_alerts` (int64), `alert_suppression_count` (int64), `hosts_compromised` (int64), `lateral_movement_hops` (int64), `persistence_mechanisms_installed` (int64), `data_volume_staged_gb` (double), `exfiltration_attempted_flag` (int64), `detection_outcome` (string), `time_to_detect_hours` (double), `mitre_tactics_observed` (string), `incident_success_flag` (int64), `defender_posture_tier` (string), `false_positive_rate_campaign` (double), `noise_signal_ratio` (double)

### `security_events.tsfile`

- **Time** (INT64, milliseconds) — the source `timestamp` timestamp as INT64 milliseconds.
- **TAG**: `host_id` — device dimension(s); query one device with `WHERE host_id=...`.
- **FIELD** (21): `event_id` (string), `user_id` (string), `source_ip` (string), `dest_ip` (string), `source_port` (int64), `dest_port` (int64), `event_class` (string), `event_type` (string), `log_source_type` (string), `severity_level` (string), `attack_lifecycle_phase` (string), `mitre_technique_id` (string), `mitre_tactic` (string), `label_malicious` (int64), `label_false_positive` (int64), `label_log_tampered` (int64), `threat_actor_id` (string), `threat_actor_profile` (string), `cvss_score_analogue` (double), `raw_log_payload` (string), `incident_id` (string)

## Conversion notes

- Each CSV of the sample is converted into its own TsFile table (4 tables), so no table is merged with another and no column is dropped.
- Event/telemetry tables use the source simulation clock as Time (INT64 milliseconds): `timestep` step counters and millisecond offsets are stored as-is, minute-based columns are multiplied by 60,000, and ISO/UTC strings are parsed to epoch milliseconds.
- Static inventory/topology/summary tables have no per-row clock; their Time is a synthetic row index (0..N-1) over the stable source order.
- Low-cardinality dimensions of each table (campaign/incident/asset/session ids, phases, labels) are declared as TAG columns; query one simulation with e.g. `WHERE campaign_id='...'`.

## Usage

Install the Apache TsFile Python SDK (`pip install tsfile`) and read a converted file:

```python
from pathlib import Path
from tsfile import TsFileReader

path = Path("alert_records.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())
```

## Source & license

- Original dataset: [`xpertsystems/cyb010-sample`](https://huggingface.co/datasets/xpertsystems/cyb010-sample)
- Source card title / publisher: CYB010 — Synthetic Security Event Log Dataset (Sample)
- License: cc-by-nc-4.0 (as declared by the original dataset).