Help Center

Complete guide to using QuantraIQ's advanced ML platform for customer analytics

Quick Start Guide

1
Upload Data

Upload your customer data files

2
Validate & Group

Validate data quality and create groups

3
Train Models

Configure and train AI models

4
Analyze Results

View insights and export reports

Platform Functions Guide

Data Management
Upload Data

Upload CSV or Excel files containing customer data. The system supports files up to 50MB with automatic format detection.

  • • Supported formats: CSV, XLSX, XLS
  • • Maximum file size: 50MB
  • • Automatic data type detection
  • • Progress tracking during upload
Data Validation

Six-stage validation pipeline ensures data quality and ML readiness with comprehensive quality scoring.

  • • File format validation
  • • Schema validation (25 required fields)
  • • Data quality assessment
  • • Business rules validation
  • • ML readiness check
  • • Automatic data transformation
Industry Templates

Download pre-configured templates for specific industries to ensure optimal data structure and validation.

  • • Insurance template (policies, claims)
  • • Retail template (transactions, products)
  • • Banking template (accounts, loans)
  • • Healthcare template (patients, treatments)
  • • Telecommunications template (services, usage)
Data Groups
Create Groups

Organize your customer data into logical groups for more precise analysis and targeted model training.

  • • Geographic segmentation
  • • Customer type classification
  • • Product category grouping
  • • Behavioral segments
Assign Data

Assign customer records to specific groups based on your business logic and analysis requirements.

  • • Bulk assignment tools
  • • Rule-based assignment
  • • Manual assignment options
  • • Group performance tracking
Group-Specific Analytics

Analyze performance metrics and train models specific to each data group for more accurate predictions.

  • • Group-specific KPIs
  • • Comparative analysis
  • • Targeted model training
  • • Segmented reporting
AI Training
Training Contexts

Define business contexts for AI-powered rule generation and model training optimization.

  • • Business objective definition
  • • Target customer type specification
  • • Industry context analysis
  • • Success metrics configuration
Intelligent Rule Generation

AI-powered suggestions for training rules based on your business objectives and customer data patterns.

  • • OpenAI-powered suggestions
  • • Domain-specific recommendations
  • • Feature engineering guidance
  • • Business context integration
Model Training

Train churn prediction and LTV models using advanced algorithms with automated hyperparameter tuning.

  • • Random Forest algorithms
  • • XGBoost implementation
  • • Automated hyperparameter tuning
  • • Cross-validation testing
Analytics & Insights
Dashboard

Comprehensive overview of your customer analytics with key performance indicators and trend analysis.

  • • Real-time KPI monitoring
  • • Customer segmentation charts
  • • Churn risk analysis
  • • Revenue trend visualization
Insights

Advanced insights including cohort analysis, customer segmentation, and performance metrics with actionable recommendations.

  • • Cohort retention analysis
  • • Customer lifetime value trends
  • • High-risk customer identification
  • • Profitability analysis
Reports

Export detailed reports and customer lists for further analysis and campaign management.

  • • High-risk customer exports
  • • High-value customer lists
  • • Performance summaries
  • • Custom report generation

Technical Support

Performance Metrics

Understanding model performance indicators and quality metrics.

  • Accuracy Score: Percentage of correct predictions
  • Precision: Reliability of positive predictions
  • Recall: Ability to identify positive cases
  • ROC AUC: Model's ability to distinguish between classes
  • R² Score: Variance explained by LTV models
Troubleshooting

Common issues and solutions for optimal platform performance.

  • Data Quality: Ensure 70%+ quality score
  • File Format: Use CSV or Excel formats
  • Required Fields: Include all 25 mandatory fields
  • Model Training: Minimum 100 records required
  • Performance: Models retrain automatically

Frequently Asked Questions

What file formats are supported?

QuantraIQ supports CSV, XLSX, and XLS file formats with automatic format detection and data type inference.

How many records do I need for training?

A minimum of 100 customer records is recommended for basic model training. For optimal performance, 1000+ records are preferred.

How often should I retrain models?

Models automatically retrain when performance degrades. Manual retraining is recommended when adding significant new data.

What is the data quality score?

A comprehensive metric (0-100) assessing data completeness, accuracy, and ML readiness. Minimum 70% required for processing.

Can I export my data and results?

Yes, export customer lists, performance reports, and analytics summaries in CSV format for further analysis.

How do I interpret ROC AUC scores?

ROC AUC measures model performance: 0.5 = random, 0.7 = good, 0.8+ = excellent. Higher scores indicate better predictive accuracy.

Need Additional Help?

Our support team is ready to assist with implementation, training, and optimization.