Gega Intelligent Document Processing(IDP)

We enable embedded Artificial Intelligence tools to capture, classify, extract and understand various types of documents, information and language coming from varied channels leveraging both supervised & unsupervised Machine Learning(ML) to build cognitive document models for document classification and data extraction –

Supervised learning

Uses learn-by-example classifiers, trainable document separation and data extraction locators to learn document type samples to automatically classify documents and extract their fields in production. We can train the models by importing real-world samples of each document type to be automated with real-time feedback.

Unsupervised learning

Uses Machine Learning(ML) to automatically organize/cluster documents in a file share by type, which accelerates cognitive model building activities during design time.

Our IDP Approach

Document Classification

  • Takes care of DATA extraction from unstructured documents
  • Enables Multi-Channel DATA Processing
  • Automated Document Classification based on content type

DATA Extraction

  • Extraction of information using business rules, artificial intelligence and fuzzy logic
  • Pre-Processing-Enhance image quality
    • Brightness
    • Contrast
    • Remove Dirt
    • Remove Lines
    • Remove Background
  • Post-Processing: Remove OCR Errors
    • Auto Correct
    • Auto Validate
    • Template-FREE Data capture

Formatting of Information

  • Formatting information in certain formats E.G. Amounts, Dates etc.

Input to Downstream Application

  • ERP
  • Workflow
  • Core Banking System
  • CRM
  • Application