Collective Intelligence Model for Education (CIME)
Tomoya Okubo
No 325, OECD Education Working Papers from OECD Publishing
Abstract:
This study introduces the Collective Intelligence Model for Evaluation (CIME), a framework designed to address the limitations inherent in traditional assessment methods. CIME integrates psychometric modelling, large language models (LLMs), and human expertise to effectively assess complex and granular competencies, including both cognitive operations and non-cognitive skills, across diverse assessment contexts. By leveraging AI-driven analysis through a continually updated knowledge database, complemented by expert oversight, CIME enhances the precision and reliability of assessments. The model establishes rigorous and continuously updated scoring criteria to ensure internal consistency, eliminate bias, and maintain international comparability. In addition, CIME incorporates advanced diagnostic methodologies, utilising psychometric scaling and Retrieval-Augmented Generation (RAG) to provide comprehensive diagnostics. This ensures that all outputs are anchored in accurate and verified sources. The study also highlights CIME’s ability to deliver personalised feedback, tailoring diagnostics to individual learners’ competencies and backgrounds. This enables personalised learning pathways and supports informed decision-making in education. Furthermore, CIME offers a scalable and efficient solution for modern educational assessment and policymaking by continuously refining its knowledge base and securely integrating with external data sources.
Date: 2025-01-29
References: Add references at CitEc
Citations:
There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:oec:eduaab:325-en
Access Statistics for this paper
More papers in OECD Education Working Papers from OECD Publishing Contact information at EDIRC.
Bibliographic data for series maintained by ().