Skip to content

Automated Detection/Conversion/Interpretation (MDMS Viewer AI)

Within the MARS Data Mining Studio (MDMS), the integrated viewer utilizes Artificial Intelligence (AI) not just for displaying content but also for preparing it for detailed data extraction and analysis. It applies intelligent processing across multiple stages.

Building upon the RTV's detection capabilities (identifying encoding, carriage control, layout), the MDMS Viewer AI adds further layers:

  • Automated Text Conversion: For non-textual sources like images or certain print streams, it invokes Optical Character Recognition (OCR) or other conversion methods automatically to generate a textual representation suitable for searching and data extraction.
  • Intelligent Interpretation for Extraction: As data is extracted based on user-defined rules or templates, the AI assists in interpreting the content within its context. This helps accurately map extracted text to the correct data types (e.g., recognizing date formats, currency symbols, addresses) even when faced with variations or ambiguities in the source. Output formats like XML or data-layered PDFs benefit from this structured interpretation.
  • Advanced OCR Capabilities: Integrates AI-powered OCR to recognize a wide range of elements from images, including standard text, specific handwriting styles, captured signatures, barcodes (various symbologies), QR codes, and checkbox/mark recognition (OMR). This significantly expands the scope of data that can be automatically digitized and mined.