What's New in Grooper 2023: Difference between revisions

From Grooper Wiki
No edit summary
Line 55: Line 55:
== GPT Integration ==
== GPT Integration ==


Version 2023 represents our first foray into GPT based extraction!  There are two new methods of using GPT API's to collect information:
Version 2023 represents our first foray into GPT based extraction!  There are two new methods of using GPT APIs to collect information:
* The '''''GPT Lookup''''' configured on a '''Data Model'''.
* The '''''GPT Lookup''''' configured on a '''Data Model'''.
* The '''''GPT Complete''''' extractor type, configurable on '''Value Readers''', '''Data Types''' and wherever extractor properties are found near you.
* The '''''GPT Complete''''' extractor type, configurable on '''Value Readers''', '''Data Types''' and wherever extractor properties are found near you.


Check back soon for an article on this topic!
Check back soon for an article on this topic!

Revision as of 15:38, 27 March 2023

Welcome to Grooper 2023!


Grooper version 2023 is here!

Below you will find brief descriptions on new and/or changed features. When available, follow any links to extended articles on a topic.

Expanded Web Client

In Grooper 2022, we introduced the Grooper Web Client for document review. In version 2023, the web client was expanded to incorporate all aspects of the Grooper product suite. After installing and configuring the Grooper Web Server, users can now design Grooper content using the web client's "Design" page, just as if they were using the thick client Grooper Design Studio application.

Instructions on installing the Grooper Web Server application and further information on how to set up your server can be found in the Install and Setup - 2023 article.

Permission Sets

Permission Sets are a new way of handling user access in Grooper 2023. Using Permission Sets, Grooper designers will define which areas of the Grooper web client users can access using Windows ACL (Access Control List) definitions.

A common Permission Sets configuration will add two sets of permissions:

  1. Designer permissions
    • Grooper admins and designers should be given full access to all pages, including "Design".
    • This will give this users full control to configure Grooper nodes in Design, as well as full rights to review Batches in the "Batches" and "Tasks" pages.
  2. Reviewer permissions
    • Grooper reviewers should have more limited access. Typically, they will only be given access to the "Batches" and/or "Tasks" pages and NEVER the "Design" page.
    • This will allow reviewers to review Batches in the "Batches" and/or "Tasks" pages, but restrict their ability to configure Grooper nodes in "Design".

Permission Sets also allow fine-grained control of object commands in ways never before capable in Grooper. For example, you can use Permission Sets to remove a document review user's ability to delete a Batch.

For more information, visit the Permission Sets article.

Data Viewer Styling with CSS

Have you ever wanted to make a Data Model look more like the document you're trying to review in a Review step? With CSS styling, now you can! In version 2023, we allow users to style how fields, tables and sections appear in the Data Viewer using CSS. This gives you unparalleled control over a Data Model's look and layout when reviewing Grooper's data extraction results.

For more information, visit the CSS Data Viewer Styling.

Web Services Lookup

Grooper can now connect to an API endpoint to validate and return data using the new Web Service Lookup found on a Data Model, much the same way you can use Grooper fields to lookup values in a database for a Database Lookup. The Web Service Lookup uses either GET or POST requests to return values from an API endpoint.

For more information, visit the Web Service Lookup article.

Improved Azure OCR Integration

Support for multiple versions of Azure Cognitive Services has been included. Users can now choose between versions 3.0, 3.1, and 3.2. If calling the Read API version 3.2 or higher, users may now select the model version as well.

We have also incorporated a new "Supplement Results" mode. This mode intends to get the best of both worlds between Azure OCR and traditional OCR. When using this mode, Transym OCR results will supplement Azure OCR results where needed. This has been demonstrated to provide more accurate OCR for densely populated table values, particularly for smaller decimal values and single character values. Furthermore, the supplemental Transym results provide character position data where Azure OCR does not. This vastly improves results where Grooper's extraction logic utilizes text position information (i.e. x,y coordinates) to collect data.

Check back soon for an article on this topic!

GPT Integration

Version 2023 represents our first foray into GPT based extraction! There are two new methods of using GPT APIs to collect information:

  • The GPT Lookup configured on a Data Model.
  • The GPT Complete extractor type, configurable on Value Readers, Data Types and wherever extractor properties are found near you.

Check back soon for an article on this topic!