Testing ChatGPT Response Variety to Introduce Natural Language Processing

This assignment, by Elisa Beshero-Bondar, is from the TextGenEd collection in the WAC Clearinghouse Repository.

The abstract from the site explains:

This sequence of assignments progressively introduces students to natural language processing (NLP) through repeated prompt experiments with ChatGPT. Students are beginners learning Python and NLP. Accessing ChatGPT and writing prompt experiments successfully provided the basis for them to investigate the cosine similarity of word embeddings in multiple responses to the same prompt. These assignments succeeded in introducing students to NLP using short generated texts prior to students’ beginning to experiment with larger text corpora.

Key Features of This Assignment

Introduction to Natural Language Processing

This assignment uses ChatGPT to introduce students to the basics of natural language processing (NLP). Students analyze how the AI processes and generates text, providing a practical understanding of NLP concepts.

Exploration of Response Variety

Students test the variety of responses generated by ChatGPT to the same prompt. This feature helps them understand the factors influencing AI-generated content and the variability in AI responses.

Critical Evaluation and Reflection

Students critically evaluate the effectiveness, relevance, and coherence of the AI-generated responses. They reflect on the potential applications and limitations of using AI in technical communication.

Find the full version of this assignment at the WAC Clearinghouse.