Skip to main content

Workbench Quick Start Guide

22/09/2025

This Quick Start Guide offers a brief introduction of the Workbench, which allows subject matter experts to easily fine-tune the settings for inference tagging and semantic recommender systems.

Inference tagging is the process of automatically identifying and tagging concepts in a text that are not explicitly mentioned but can be inferred from context, leveraging a knowledge graph and user-defined rules. The semantic recommender system, in contrast, uses this enriched, semantically-tagged content to generate personalized document suggestions, leading to more relevant and context-aware recommendations.

The Workbench supports subject matter experts in finding an optimal combination of parameters for inference tagging and semantic recommender systems.

After login, you will see the main screen where you can provide the input text (1) and click the Arrow icon (2) to get results based on your input. You can then use the Inference Tagging (3) and Recommender (4) tabs to verify and test your Workbench configurations.

workbench_landingpage_9_7_1.png
workbench_inference_example.png
workbench_recommender_example.png

You can also change the default parameters for the process of semantic extraction, expansion and recommendation manually or in bulk by importing a previously exported configuration.

Once you have found an optimal combination of parameters, you can export them as cURL requests. Another option is to save the settings as a configuration into the application's database and thus make it accessible to other users who have access to your Workbench instance.