Optimizing Routine Educational Tasks through Prompt Engineering: A Comparative Study of AI Chatbots
The rapid integration of artificial intelligence (AI) in education necessitates the development of effective strategies for optimizing routine teaching tasks. This study explores the relevance of prompt engineering as a tool for enhancing AI-generated responses, reducing educators' workload, an...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Anhalt University of Applied Sciences
2025-04-01
|
| Series: | Proceedings of the International Conference on Applied Innovations in IT |
| Subjects: | |
| Online Access: | https://icaiit.org/paper.php?paper=13th_ICAIIT_1/1_5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The rapid integration of artificial intelligence (AI) in education necessitates the development of effective strategies for optimizing routine teaching tasks. This study explores the relevance of prompt engineering as a tool for enhancing AI-generated responses, reducing educators' workload, and improving the efficiency of lesson planning, content creation, and assessment design. The primary objective of this research is to develop and evaluate a structured methodology for designing prompts that maximize the relevance, completeness, and applicability of AI-generated outputs. To achieve this goal, a three-phase methodology was employed: (1) a preparatory phase involving a literature review and the development of standardized educational prompts, (2) an experimental phase testing these prompts across multiple AI chatbot models (Claude, GPT, and Copilot), and (3) an analytical phase assessing chatbot responses based on predefined criteria, including relevance, accuracy, completeness, practicality, and structuredness. The results indicate significant differences in chatbot performance. Claude demonstrated superior contextual understanding, GPT provided well-balanced and structured responses, while Copilot exhibited high factual accuracy but required improvements in contextual adaptation. Statistical analysis using the Kruskal-Wallis H test confirmed these variations, highlighting the necessity of model-specific prompt optimization. The study’s findings have both practical and theoretical significance. Practically, they provide educators with a structured approach to prompt engineering, enabling more effective use of AI tools in teaching. Theoretically, the research contributes to the growing field of AI- assisted education by offering insights into optimizing human-AI interaction. The conclusions emphasize the need for continued refinement of AI models and further exploration of prompt engineering techniques. Future research should focus on expanding testing across various disciplines and integrating AI-driven tools into digital learning environments to enhance personalized education. |
|---|---|
| ISSN: | 2199-8876 |