Objective: To conduct a comprehensive analysis of AI-based personalized math learning platforms for K-12 education using Montebello's (2018) framework.

Methods:

• Analyzed 12 global math learning platforms using criteria based on Montebello's framework of Personal Learning Portfolios (PLP), Personal Learning Networks (PLN), and Personal Learning Environments (PLE)
• Evaluated platforms' features and functionalities related to personalized learning

Key Findings:

• Most platforms aligned well with PLP through student/teacher dashboards and reward systems
• PLN features were limited, with diverse learning materials but few peer interaction opportunities
• PLE features were common, especially adaptive content based on diagnostic assessments
• Platforms lacked balanced implementation across all three framework components

Implications:

• Provides insights for effective integration of AI-driven technologies in math education
• Highlights need for more balanced design incorporating peer interaction features
• Suggests empirical verification of design principles' effectiveness is needed

Limitations:

• Focus on specific set of 12 platforms may not represent all available options
• Analysis based on reported features rather than direct testing of platforms

Future Directions:

• Design AI-based platforms that facilitate comprehensive, balanced learning across PLP, PLN, and PLE
• Empirically verify effectiveness of specific design principles for different learners and outcomes

Title and Authors: "Comprehensive Analysis of AI-based Math Learning Platforms for K-12 Education" by Seonghye Yoon, Soyeon Min, and Daeun Kang

Published On: October 7-10, 2024 Published By: eLearn 2024 Conference Proceedings, Singapore