May 2022


Intelligent Document Processing Fully Functional

We are very pleased to announce that our IDP suite below are now out of the beta phase and fully ready for production use.

Knowledge Graphs Available for Fixed Form Extractors

A Knowledge Graph is an Automation Hero Context Aware feature for our intelligent document processes. Enabling a Knowledge Graph makes Automation Hero's OCR engines look at the possible expected field values in your configured Knowledge Graph to make better predictions when reading a field. This results in higher field-level accuracy than a standard dictionary lookup or validation methods. 

Adding a Knowledge Graph is also quite simple. Either import the data from one of your Hero Platfrom_ Connections or import it directly to the Data Store. Once your information is in the Data Store it can be enabled as a Knowledge Graph and used in the Fixed Form Data Extractions to help improve OCR results.

Fixed Form Processing Improvements

The Fixed Form Document Extractor has been streamlined to process your data faster than before. We found new methods to improve calculation times when creating your models in the Document Studio.

AI Page Identification 

AI page identification is a new feature that checks if input images used in a Fixed Form Document Extractor function match the order of the training images created in the Document Studio.

Turning on the AI page identification features allows for processing input images in a Flow even if they don't match the order and number of the sample images from the Document Studio model. 

 

Multiple Languages Supported for Typed Fields

The Fixed Form Document Extractor and the Custom Invoice Extractor now support a wide variety of different languages that our AI can recognize in your typed text fields. You can now select the language for your typed text which gives the AI engine a high probability to correctly read values in your documents. 

Supported languages:

  • Arabic, Bulgarian, Czech, German, Greek, English, French, Hindi, Croatian, Hungarian, Indonesian, Italian, Japanese, Dutch/Flemish, Norwegian, Polish, Portuguese, Russian, Slovak, Slovenian, Spanish/Castilian, Turkish, Ukrainian

Improved Bounding Boxes for Labeling Data

We have updated the feature for drawing bounding boxes the Human-in-the-Loop application of Hero Platfrom_. This feature has been improved so that boxes appear in the same location on your documents, even if your documents have changed in size (DPI). Our List Page Bounding Boxes also allow for multiple locations to be to be selected when a value you want to capture appears multiple times on your documents.

New Function: Repartition

The Repartition function creates partitions from input data.

Some input data (e.g., a large CSV file) does not allow for partitioning and cannot be set to run distributed. The Repartition function can overcome this obstacle. When data passes through the Repartition function, it is stored on the local hard disk and added into multiple partitions depending on the parallelism configured in the Flow.

Status Bar Environment Metrics

The status bar appears at the bottom of the user interface throughout Hero Platfrom_ . It can be helpful to monitor your environment's resources and plan your automations accordingly. 

Resources:

  • JVM memory
  • Machine memory
  • CPU
  • Jobs: Running and queued




Search