Documentation
Explore our guides and examples to integrate GatekeeperX.
Introduction to Payment Fraud and GatekeeperX
Technical overview of the payment fraud mitigation system.
Defining Payment Fraud
Payment fraud in the online domain encompasses the unauthorized utilization of compromised payment card data for illicit transactions. Such data, frequently sourced from underground marketplaces, facilitates fraudulent purchases and subsequent reselling for financial gain.
GatekeeperX: Real-time Fraud Assessment
GatekeeperX provides a robust, API-driven platform for mitigating payment fraud. Clients integrate by transmitting comprehensive order telemetry via API endpoints. GatekeeperX leverages advanced analytics to perform real-time fraud assessments, returning definitive accept/reject recommendations.
Key Technologies
- Machine Learning models trained with client-specific transaction data
- Advanced rule-based engine for strict business policy enforcement
- Graph network analysis for detecting entity interconnections
- RESTful API integration for real-time transaction assessment
Quick Start
Get started with GatekeeperX in minutes with this quick guide.
1Get API Credentials
To get started, you'll need to request your API credentials. Contact our support team to obtain your API Key and Secret.
Example Credentials
# Your credentials will look similar to this
API_KEY="gkx_live_xxxxxxxxxxxxxxxxxxxxxxxx"
API_SECRET="gkx_secret_xxxxxxxxxxxxxxxxxxxxxxxx"2Set Up Your Environment
Configure your environment with the provided credentials and select the appropriate endpoint for your region.
Configuration File
{
"environment": "production",
"apiKey": "gkx_live_xxxxxxxxxxxxxxxxxxxxxxxx",
"apiEndpoint": "https://api.gatekeeperx.com/v1",
"timeout": 30000,
"webhookSecret": "wh_secret_xxxxxxxxxxxxxxxxxxxxxxxx"
}3Perform Your First Risk Assessment
Once configured, you can perform your first risk assessment by making a request to our API.
Assessment Request with JavaScript
const GatekeeperX = require('gatekeeperx');
const gkx = new GatekeeperX({
apiKey: process.env.GKX_API_KEY
});
// Perform a risk assessment
async function evaluateRisk() {
try {
const assessment = await gkx.riskAssessments.create({
eventType: "login",
userId: "user_123",
ipAddress: "192.168.1.1",
userAgent: "Mozilla/5.0..."
});
console.log(assessment.riskScore);
console.log(assessment.riskLevel);
console.log(assessment.triggers);
} catch (error) {
console.error("Error:", error);
}
}
evaluateRisk();Core Technologies Driving Fraud Detection
GatekeeperX utilizes a multi-layered detection approach.
Machine Learning (ML) Models
Custom Model Training: ML models are built per client, leveraging client-specific transaction data for optimized detection.
Adaptive and Scalable: ML models are dynamic, adapting to evolving fraud patterns, ensuring scalability and resistance to static evasion techniques.
Model Lifecycle: Continuous monitoring of performance, retraining, and feature set optimization.
Feature Engineering
Models ingest hundreds of features:
Transaction/Order History: Temporal analysis of transaction frequency, order volumes, and spending patterns.
Geospatial Data: Address discrepancies (billing vs. shipping), IP geolocation.
Device Fingerprinting: Device IDs, metadata, and historical usage patterns.
Behavioral Analysis: Checkout interaction analysis, including user journeys and mouse movements.
Customer Attributes: Email domains, account age, and registration details.
Merchant Attributes: Bank account, tenure, historical volumes, domain, and shipping patterns.
Rule-Based Engine
Business Logic Enforcement: Rules allow strict enforcement of business policies and rapid deployment of safeguards.
Flexibility:
- documentation.fraudDetection.ruleBasedEngine.examples[0]
- documentation.fraudDetection.ruleBasedEngine.examples[1]
- documentation.fraudDetection.ruleBasedEngine.examples[2]
- documentation.fraudDetection.ruleBasedEngine.examples[3]
Hybrid Model: Rules can augment ML models (e.g., dynamic 3DS activation based on ML score + device heuristics).
Continuous Optimization: GatekeeperX provides AI agents that automate rule performance feedback and suggest improvements.
Graph Network Analysis
Entity Interconnection: Real-time analysis of linkages between customers, devices, and payment methods.
Anomaly Detection:
- Rapid Network Expansion: Detection of explosive growth patterns indicating coordinated fraud.
- Resource Sharing Anomalies: Detection of excessive sharing of payment cards, devices, or emails.
- Chargeback Propagation: Identification of networks associated with confirmed chargeback events.
ML Reinforcement: Graph features (connectivity metrics) are used as additional features for ML models.
Risk Assessment
How risk levels are scored and evaluated.
Scoring Models
GatekeeperX uses sophisticated scoring models to evaluate risk:
documentation.riskAssessment.scoringModels.diagram.title
documentation.riskAssessment.scoringModels.example.title
documentation.riskAssessment.scoringModels.example.codedocumentation.riskAssessment.riskFactors.title
documentation.riskAssessment.riskFactors.description
- 1
documentation.riskAssessment.riskFactors.factor1.title
documentation.riskAssessment.riskFactors.factor1.description
- 2
documentation.riskAssessment.riskFactors.factor2.title
documentation.riskAssessment.riskFactors.factor2.description
- 3
documentation.riskAssessment.riskFactors.factor3.title
documentation.riskAssessment.riskFactors.factor3.description
- 4
documentation.riskAssessment.riskFactors.factor4.title
documentation.riskAssessment.riskFactors.factor4.description
documentation.riskAssessment.riskFactors.customization.title
documentation.riskAssessment.riskFactors.customization.description
API Integration Methodology
GatekeeperX offers straightforward and secure API integrations.
Core API Endpoints
Discover the essential endpoints for integration with GatekeeperX:
Risk Scoring
Evaluate transactions in real-time
User Analysis
User profiling and behavior analysis
Device Verification
Device identification and validation
Integration Process
Follow these steps for successful integration:
API Account Setup
Endpoint Implementation
Integration Testing
Production Deployment
Continuous Monitoring
Go-Live Process
Steps for launching GatekeeperX in production:
Example Requests
Code examples for quick integration:
// Example transaction assessment request
const response = await fetch('https://api.gatekeeperx.com/v1/assessments', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
'Authorization': 'Bearer YOUR_API_KEY'
},
body: JSON.stringify({
event_type: 'payment',
transaction_id: 'tx_123456789',
amount: 299.99,
currency: 'USD',
customer: {
id: 'cust_987654321',
email: 'customer@example.com',
ip_address: '192.168.1.1',
device_id: 'dev_abcdef123456'
},
payment: {
method: 'credit_card',
bin: '411111',
last4: '1234'
}
})
});
const result = await response.json();
console.log(result.risk_score); // 85
console.log(result.risk_level); // high
console.log(result.recommendation); // rejectBest Practices
Recommendations for optimal integration:
- Send enriched data for improved accuracy
- Implement error response handling
- Use appropriate timeouts
- Regularly monitor performance
- Keep API keys secure