3/15/2022

KNIME: GPT3 component

KNIME: GPT3 component

As many of you know, lately I am focusing on the no-code / low-code KNIME platform to explore flexible solutions to both data processing (ETL) and natural language processing (NLP).

Just an intro to GPT3

GPT3 is one of the most powerful language models developed by OpenAI. In this article, we explore how to integrate it into KNIME so that non-technical profiles can take advantage of its capabilities.

The workflow

After some iterations, this is the proposed workflow with the different stages:

KNIME Workflow

  1. S2 API Search: Retrieving papers from Semantic Scholar.
  2. Clustering: Applying clustering techniques.
  3. GPT3 Info Completion: Generating descriptions for the clusters.

GPT3 Info Completion component

This component is the heart of the integration. It allows you to configure GPT3 parameters directly from the KNIME interface:

GPT3 Configuration

Results

The results provide a clear and descriptive view of the topics covered in a set of documents, greatly facilitating bibliographic analysis.

Clustering Results

You can download the full workflow from the KNIME Hub.