[Paper Review] Tailoring Education with GenAI: A New Horizon in Lesson Planning

Kostas Karpouzis, Dimitris Pantazatos+2 more|arXiv (Cornell University)|Feb 12, 2024

Open Education and E-Learning7 citations

TL;DR

The paper presents a GenAI-based Learning Scenario Assistant that uses interactive mega-prompts to tailor lesson plans to diverse learners, with a mixed-method evaluation across languages and educational levels.

ABSTRACT

The advent of Generative AI (GenAI) in education presents a transformative approach to traditional teaching methodologies, which often overlook the diverse needs of individual students. This study introduces a GenAI tool, based on advanced natural language processing, designed as a digital assistant for educators, enabling the creation of customized lesson plans. The tool utilizes an innovative feature termed 'interactive mega-prompt,' a comprehensive query system that allows educators to input detailed classroom specifics such as student demographics, learning objectives, and preferred teaching styles. This input is then processed by the GenAI to generate tailored lesson plans. To evaluate the tool's effectiveness, a comprehensive methodology incorporating both quantitative (i.e., % of time savings) and qualitative (i.e., user satisfaction) criteria was implemented, spanning various subjects and educational levels, with continuous feedback collected from educators through a structured evaluation form. Preliminary results show that educators find the GenAI-generated lesson plans effective, significantly reducing lesson planning time and enhancing the learning experience by accommodating diverse student needs. This AI-driven approach signifies a paradigm shift in education, suggesting its potential applicability in broader educational contexts, including special education needs (SEN), where individualized attention and specific learning aids are paramount

Motivation & Objective

• Motivate personalized education through GenAI-driven lesson planning.
• Develop an interactive prompt methodology to customize learning scenarios.
• Evaluate the effectiveness of GenAI-generated lesson plans via quantitative and qualitative metrics.

Proposed method

• Introduce an interactive two-way prompt workflow to guide GenAI in lesson plan creation.
• Define positions prompts, interactive prompts, and follow-up prompts to collect detailed user requirements.
• Iteratively develop and evaluate lesson plans with end-user feedback.
• Apply linguistic and thematic analyses (spaCy, LDA) to assess generated content.
• Test across multiple LLMs (ChatGPT 3.5, ChatGPT 4, Llama variants, Google Bard) and languages (English, Greek).

Experimental results

Research questions

• RQ1 How effective is GenAI in generating tailored lesson plans that meet diverse learner needs?
• RQ2 Can an interactive prompt framework produce high-quality, standards-aligned educational content across subjects and levels?
• RQ3 What are the differences in GenAI performance across languages and models for educational content generation?

Key findings

• GenAI-generated lesson plans significantly reduce planning time and accommodate diverse needs.
• The Interactive Prompt methodology enables a dialogic, user-centered design process for lesson planning.
• Cross-model evaluations show performance varies by model and language, with newer models (ChatGPT 4, Llama 70B) generally scoring higher on relevance and personalization in English and Greek.
• Linguistic analysis (spaCy, LDA) provides insight into content quality and thematic coverage across responses.
• There is variability in resource accuracy (e.g., non-existent links) among some models, highlighting the need for human verification.