An Ontology for Social Determinants of Education (SDoEd) based on Human-AI Collaborative Approach
Project Overview
The document explores the application of generative AI, specifically ChatGPT-3.5, in creating a Social Determinants of Education (SDoEd) ontology that identifies and validates concepts related to how social factors influence educational outcomes. This ontology serves as a standardized framework to analyze the relationship between students' life circumstances and their academic performance, highlighting the crucial role education plays in mitigating inequalities and enhancing health outcomes. It comprises 231 domain concepts that have been rigorously validated through expert reviews and software tools. The findings underscore the potential of generative AI in advancing educational research and practice by providing a structured approach to understanding the various social determinants affecting students, ultimately aiming to improve educational systems and reduce disparities.
Key Applications
Social Determinants of Education (SDoEd) ontology
Context: Developed for academic research and educational policy analysis, targeting educators, researchers, and policymakers.
Implementation: Utilized ChatGPT-3.5 for concept extraction and validation, then implemented using Protégé 5.5 as an OWL file.
Outcomes: Created a comprehensive framework for understanding how social determinants affect educational outcomes, facilitating better analysis and decision-making.
Challenges: Ensuring the comprehensiveness and accuracy of the ontology, validating the relationships between concepts.
Implementation Barriers
Technical barrier
Challenges in validating the accuracy and comprehensiveness of the ontology concepts.
Proposed Solutions: Utilized both forward and backward validation methods, cross-referencing with peer-reviewed articles and employing human expert evaluators.
Project Team
Navya Martin Kollapally
Researcher
James Geller
Researcher
Patricia Morreale
Researcher
Daehan Kwak
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Navya Martin Kollapally, James Geller, Patricia Morreale, Daehan Kwak
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai