Extract Entity

This document is a step by step guide to identify and classify key entities — such as people, organizations, locations, products, and custom business terms in a document.

  1. From the top navigation choose Knowledge GraphAddExtract Entity.

  2. The Extract Entity wizard opens (Basic Configuration). Provide Task name and select the LLM to use.

3.2 Attach prompts to extraction

  1. Under Prompts in the Basic Configuration step you will see:

  • Options to Create new prompt or Add from library — use either to associate prompts with your extraction task.

  1. Add from library: search and select the prompt(s) you want, then click Add prompts. The selected prompt(s) appear as “Additional Prompts (N)”.

  2. LLM Selection: choose the model that will execute the prompts (consider capability & cost).

3.3 Source & Target dataset selection

  1. Source — pick the dataset/connection to read raw documents from. The UI provides a searchable tree.

  2. Target — specify Workspace, Collection, and Directory where extracted results will be written.

  3. Click Next to configure advanced options (embeddings, vector DB) and schedule.

  4. Task Scheduling

    • Recurring job - turn it on and choose frequency at which the job should run

    • Advance settings - choose appropriate "Degree of Parallelism" and "No. of cores" for spark to parallelize processing of the data

    Scheduling step: choose One-time or toggle Recurring Job and provide cadence if recurring. Confirm and create the task. Image (schedule & summary):

  5. Task Summary

    • Review the details

    • Click " Create Task" if everything looks okay

Last updated