README.md

---
license: cc-by-nc-4.0
task_categories:
- tabular-classification
- time-series-forecasting
tags:
- tsfile
- timeseries
- cybersecurity
- identity-security
- account-takeover
- mfa-bypass
- ueba
- zero-trust
- apt
- synthetic-data
- lateral-movement
- golden-ticket
- time-series
- format:tsfile
pretty_name: CYB006 — Synthetic Login Activity Dataset (Sample) (TsFile)
size_categories:
- 1K<n<10K
configs:
- config_name: default
  data_files:
  - split: attack_timelines
    path: attack_timelines.tsfile
  - split: auth_events
    path: auth_events.tsfile
  - split: campaign_events
    path: campaign_events.tsfile
  - split: campaign_summary
    path: campaign_summary.tsfile
  - split: identity_topology
    path: identity_topology.tsfile
  - split: login_sessions
    path: login_sessions.tsfile
  - split: user_risk_summary
    path: user_risk_summary.tsfile
  - split: victim_topology
    path: victim_topology.tsfile
---

# CYB006 — Synthetic Login Activity Dataset (Sample) (TsFile)

Apache TsFile version of [`xpertsystems/cyb006-sample`](https://huggingface.co/datasets/xpertsystems/cyb006-sample).

## Overview

**CYB006 — Synthetic Login Activity Dataset (Sample)** — This is a **free preview** of the full **CYB006 — Synthetic Login Activity Dataset** product. It contains roughly **~1.3% of the full dataset** at identical schema, threat-actor-tier distribution, and statistical fingerprint, so you can evaluate fit before licensing the full product.

- **Tables:** `attack_timelines`, `auth_events`, `campaign_events`, `campaign_summary`, `identity_topology`, `login_sessions`, `user_risk_summary`, `victim_topology`
- **Rows:** 265,676 across 8 TsFile table(s).
  - `attack_timelines`: 37,489 rows — Time is the source `timestep` simulation step counter, stored as milliseconds (1 step = 1 ms).
  - `auth_events`: 31,900 rows — Time is the source `event_timestamp_utc` second counter on the simulation clock, converted to milliseconds.
  - `campaign_events`: 190,137 rows — Time is the source `timestep` simulation step counter, stored as milliseconds (1 step = 1 ms).
  - `campaign_summary`: 500 rows — Time is synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
  - `identity_topology`: 150 rows — Time is synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
  - `login_sessions`: 5,000 rows — Time is the source `session_timestamp_utc` second counter on the simulation clock, converted to milliseconds.
  - `user_risk_summary`: 200 rows — Time is synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
  - `victim_topology`: 300 rows — Time is synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.

## Schema (TsFile structure)

### `attack_timelines.tsfile`

- **Time** (INT64, milliseconds) — the source `timestep` simulation step counter, stored as milliseconds (1 step = 1 ms).
- **TAG**: `campaign_id` — device dimension(s); query one device with `WHERE campaign_id=...`.
- **FIELD** (19): `actor_id` (string), `attack_phase` (string), `files_encrypted_cumulative` (int64), `encryption_throughput_mbps` (double), `endpoints_compromised` (int64), `lateral_move_count` (int64), `credential_harvest_count` (int64), `c2_bytes_exfiltrated` (double), `defender_alert_score` (double), `detection_outcome` (string), `blast_radius_pct` (double), `actor_capability_tier` (string), `living_off_land_score` (double), `attribution_risk_score` (double), `data_exfiltrated_gb` (double), `wiper_flag` (int64), `double_extortion_flag` (int64), `ir_activated` (int64), `target_segment_id` (string)

### `auth_events.tsfile`

- **Time** (INT64, milliseconds) — the source `event_timestamp_utc` second counter on the simulation clock, converted to milliseconds.
- **TAG**: `user_id` — device dimension(s); query one device with `WHERE user_id=...`.
- **FIELD** (8): `event_id` (string), `session_id` (string), `event_type` (string), `target_account_id` (string), `source_domain_id` (string), `target_domain_id` (string), `application_id` (string), `risk_score_delta` (double)

### `campaign_events.tsfile`

- **Time** (INT64, milliseconds) — the source `timestep` simulation step counter, stored as milliseconds (1 step = 1 ms).
- **TAG**: `campaign_id` — device dimension(s); query one device with `WHERE campaign_id=...`.
- **FIELD** (6): `actor_id` (string), `event_type` (string), `target_segment_id` (string), `defender_id` (string), `affected_endpoint_id` (string), `event_detail` (string)

### `campaign_summary.tsfile`

- **Time** (INT64, milliseconds) — synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
- **TAG**: `campaign_id` — device dimension(s); query one device with `WHERE campaign_id=...`.
- **FIELD** (42): `actor_id` (string), `actor_capability_tier` (string), `ransomware_family` (string), `operator_model` (string), `victim_industry_vertical` (string), `target_segment_id` (string), `backup_maturity_tier` (string), `soc_maturity_tier` (string), `edr_coverage_rate` (double), `network_segmentation_quality` (double), `total_dwell_time_hrs` (double), `endpoints_encrypted_total` (int64), `blast_radius_pct` (double), `files_encrypted_total` (int64), `data_exfiltrated_gb` (double), `ransom_demand_usd` (double), `ransom_paid_flag` (int64), `recovery_time_hrs` (double), `total_business_downtime_hrs` (double), `ir_containment_speed_score` (double), `backup_recovery_success_flag` (int64), `double_extortion_flag` (int64), `wiper_component_flag` (int64), `lateral_move_count_total` (int64), `credential_harvest_count_total` (int64), `detection_phase` (string), `campaign_success_flag` (int64), `financial_impact_score` (double), `resilience_score` (double), `mean_time_to_detect_hrs` (double), `mean_time_to_contain_hrs` (double), `edr_alerts_total` (int64), `siem_fires_total` (int64), `vss_deletion_success` (int64), `backup_destruction` (int64), `worm_spread_flag` (int64), `coordinated_flag` (int64), `living_off_land_score` (double), `lateral_coverage_ratio` (double), `encrypted_data_recoverability` (double), `insurance_loss_ratio_score` (double), `attack_efficiency_score` (double)

### `identity_topology.tsfile`

- **Time** (INT64, milliseconds) — synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
- **TAG**: `domain_id` — device dimension(s); query one device with `WHERE domain_id=...`.
- **FIELD** (13): `application_id` (string), `domain_type` (string), `trust_relationship_type` (string), `mfa_enforcement_policy` (string), `conditional_access_strength` (double), `user_count` (int64), `privileged_account_ratio` (double), `service_account_count` (int64), `ueba_coverage_level` (double), `directory_sync_enabled` (int64), `sso_provider` (string), `session_timeout_seconds` (int64), `password_policy_strength` (double)

### `login_sessions.tsfile`

- **Time** (INT64, milliseconds) — the source `session_timestamp_utc` second counter on the simulation clock, converted to milliseconds.
- **TAG**: `user_id` — device dimension(s); query one device with `WHERE user_id=...`.
- **FIELD** (20): `session_id` (string), `session_phase` (string), `login_outcome` (string), `source_ip_hash` (string), `geo_country_code` (string), `geo_city_hash` (string), `device_id_hash` (string), `device_trust_level` (string), `authentication_method` (string), `mfa_challenge_type` (string), `mfa_response_latency_ms` (double), `credential_attempt_count` (int64), `session_duration_seconds` (int64), `target_application_id` (string), `privilege_level_accessed` (string), `user_risk_tier` (string), `threat_actor_capability_tier` (string), `geo_anomaly_score` (double), `velocity_anomaly_score` (double), `impossible_travel_flag` (int64)

### `user_risk_summary.tsfile`

- **Time** (INT64, milliseconds) — synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
- **TAG**: `user_id` — device dimension(s); query one device with `WHERE user_id=...`.
- **FIELD** (19): `user_risk_tier` (string), `total_login_attempts` (int64), `successful_logins` (int64), `failed_logins` (int64), `mfa_failures` (int64), `impossible_travel_events` (int64), `lateral_hop_count` (int64), `privilege_escalations` (int64), `account_lockout_count` (int64), `geo_dispersion_score` (double), `login_velocity_score` (double), `session_anomaly_rate` (double), `ueba_alert_count` (int64), `threat_actor_flag` (int64), `account_takeover_flag` (int64), `overall_identity_risk_score` (double), `peak_privilege_level_accessed` (string), `credential_attack_victim_flag` (int64), `insider_threat_indicator_score` (double)

### `victim_topology.tsfile`

- **Time** (INT64, milliseconds) — synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
- **TAG**: `segment_id` — device dimension(s); query one device with `WHERE segment_id=...`.
- **FIELD** (14): `defender_id` (string), `segment_type` (string), `soc_maturity_tier` (string), `backup_maturity_tier` (string), `edr_coverage_rate` (double), `network_segmentation_quality` (double), `patch_posture_score` (double), `ir_activation_latency_hrs` (double), `endpoint_count` (int64), `ad_domain_complexity` (double), `soc_maturity_score` (double), `backup_recovery_prob` (double), `backup_recovery_hrs_mean` (double), `siem_rule_refresh_cadence_days` (int64)

## Conversion notes

- Each CSV of the sample is converted into its own TsFile table (8 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("attack_timelines.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/cyb006-sample`](https://huggingface.co/datasets/xpertsystems/cyb006-sample)
- Source card title / publisher: CYB006 — Synthetic Login Activity Dataset (Sample)
- License: cc-by-nc-4.0 (as declared by the original dataset).