Skip to main content Skip to navigation

VizGroup: An AI-Assisted Event-Driven System for Real-Time Collaborative Programming Learning Analytics

Project Overview

The document explores the implementation of VizGroup, an innovative AI-assisted system designed to enhance programming education through the use of Large Language Models (LLMs). By offering real-time collaborative learning analytics, VizGroup enables instructors to effectively monitor student collaboration and performance during peer instruction activities. Key applications of this system include real-time data visualization that tracks collaboration metrics and provides intelligent notifications, which collectively assist educators in managing large programming classes more efficiently. The findings indicate that the integration of generative AI in education not only improves student engagement and understanding but also facilitates better instructional strategies tailored to individual and group learning dynamics. Overall, the document highlights the transformative potential of AI technologies in fostering a more interactive and responsive educational environment, ultimately aiming to enhance learning outcomes in programming education.

Key Applications

VizGroup, an AI-assisted system for real-time collaborative programming learning analytics.

Context: Used in large programming classes (e.g., over 100 students) during Peer Instruction sessions.

Implementation: Implemented as a web-based tool utilizing LLMs to analyze collaboration dynamics and provide real-time visualizations and notifications.

Outcomes: Enhanced instructors' ability to monitor group dynamics, track student performance, and facilitate timely interventions, leading to improved engagement and learning outcomes.

Challenges: Instructors faced cognitive overload in tracking multiple patterns; the system required careful management of notifications to avoid information fatigue.

Implementation Barriers

Cognitive Overload

Instructors may struggle to keep track of numerous notifications and patterns simultaneously, leading to potential oversight of critical student issues.

Proposed Solutions: Implementing context-aware notifications and allowing customization of alerts to focus on relevant metrics.

System Complexity

The complexity of the system may overwhelm instructors, particularly those less familiar with technology.

Proposed Solutions: Providing thorough training and support resources to help instructors navigate the tool effectively.

Project Team

Xiaohang Tang

Researcher

Sam Wong

Researcher

Kevin Pu

Researcher

Xi Chen

Researcher

Yalong Yang

Researcher

Yan Chen

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Xiaohang Tang, Sam Wong, Kevin Pu, Xi Chen, Yalong Yang, Yan Chen

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

Let us know you agree to cookies