Professional-grade synthetic dataset generator for cybersecurity research and security testing
This tool generates classified environment datasets specifically designed for insider threat research in secure organizations. Unlike generic employee data generators, this system models classified document handling, security clearance levels, and sensitive operational activities that are critical for understanding insider threats in government agencies, defense contractors, and high-security enterprises.
- Multi-Level Security Classifications (Levels 1-4) for employees and documents
- Classified Document Burning simulation with security level tracking
- Clearance-Based Access Patterns reflecting real security protocols
- Hostile Country Travel tracking with security implications
- Cross-Classification Activities modeling unauthorized access scenarios
# Basic dataset generation
python main.py --employees 1000 --days 180
# With noise injection for realism
python main.py --employees 500 --days 90 --add-noise --malicious-ratio 0.08
# Analysis only mode
python main.py --analysis-only --input-file existing_data.csv- Employee Profiles → Creates realistic employee profiles with departments, roles, and characteristics
- Behavioral Simulation → Generates daily activities (building access, printing, document burning, travel)
- Analysis & Insights → Performs comprehensive behavioral and security analysis
- Export & Documentation → Outputs data in CSV/Excel with detailed documentation
The generator creates classified environment datasets with 40+ security-focused features:
employee_id,date,employee_classification,is_malicious,avg_request_classification,max_request_classification,num_burn_requests,hostile_country_trip,...
001,2024-01-01,3,0,2.5,4,2,0,...
002,2024-01-01,4,1,3.8,4,5,1,...
- Employee Security Clearance (Levels 1-4)
- Document Classification Levels (average and maximum per activity)
- Classified Document Destruction with volume and classification tracking
- Security Risk Indicators including hostile country travel and cross-classification access
- Clearance-Based Activity Patterns reflecting security protocol compliance
- Government Agency Studies: Insider threat patterns in classified environments
- Defense Contractor Analysis: Security clearance-based behavioral modeling
- Intelligence Community Research: Multi-level security access pattern analysis
- Clearance Violation Detection: Training algorithms to detect unauthorized access
- Classification Leak Prevention: Modeling document exfiltration scenarios
- Security Protocol Testing: Validating clearance-based access controls
- Insider Threat Training: Realistic scenarios for security awareness programs
- Security Audit Preparation: Test data for classified environment audits
- Compliance Testing: DCID 6/3, ICD 503, and other security standard validation
- Risk Assessment Models: Development of clearance-based risk scoring systems
- Employee Generator - Creates realistic employee profiles and organizational structures
- Activity Generators - Simulates daily employee activities and behaviors
- Analyzers - Performs behavioral and security analysis on generated data
- Data Export - Exports datasets with comprehensive documentation
- Configuration - Manages behavioral patterns and organizational settings
- Core Infrastructure - Workflow management and system utilities
- CLI Interface - Command-line tools and user interface
- 🏗️ Technical Architecture Overview - System design and data flow
- 📖 User Guide & Quick Reference - Practical usage guide with examples
- 📊 Dataset Schema & Field Definitions - Complete dataset structure documentation
- Complete Usage Guide - Detailed command-line options and examples
- Configuration Guide - Customizing behavioral patterns and settings
- Analysis Features - Understanding analysis outputs and metrics
- Technical Architecture - System design and workflow details
- Python 3.8+
- pandas, numpy, openpyxl
- 50MB+ free disk space
- 1GB+ RAM for large datasets (5000+ employees)
MIT License - See LICENSE for details.
Ready to generate your first insider threat dataset? Start with python main.py --help for full options.