DataliVR: Transformation of Data Literacy Education through Virtual Reality with ChatGPT-Powered Enhancements
Project Overview
The document examines the development and implementation of DataliVR, a virtual reality application aimed at improving data literacy among university students through the integration of a ChatGPT-powered AI chatbot. It underscores the critical importance of data literacy in the digital age and evaluates the effectiveness of combining VR and AI in educational contexts. While findings indicate that the inclusion of the chatbot enhanced user experience and system usability, it did not lead to a significant improvement in learning outcomes when compared to scenarios without the chatbot. The research highlights the necessity for further investigation into the role of AI in educational VR environments to better understand its potential impact on learning effectiveness.
Key Applications
DataliVR - a VR application for data literacy education enhanced with a ChatGPT-powered chatbot.
Context: Higher education, targeting university students with varied academic backgrounds but no prior knowledge of machine learning.
Implementation: Implemented as a gamified VR environment with stages for data collection, cleaning, analysis, and visualization, using a ChatGPT API for personalized assistance.
Outcomes: Enhanced user experience and system usability; high engagement in data literacy tasks; positive feedback on the chatbot's effectiveness.
Challenges: The presence of the chatbot did not significantly improve learning outcomes; user engagement with the chatbot was relatively low.
Implementation Barriers
Technical barrier
Participants expressed a higher need for technical support when using DataliVR with the chatbot compared to without. Future designs should incorporate more intuitive interactions with the chatbot to reduce the perceived need for technical assistance.
Proposed Solutions: Incorporate more intuitive interactions with the chatbot to reduce the perceived need for technical assistance.
Engagement barrier
Participants had low interaction rates with the chatbot, indicating a lack of engagement during the tasks. Encourage more active user engagement with the chatbot during the learning process by presenting knowledge-related challenges.
Proposed Solutions: Encourage more active user engagement with the chatbot during the learning process by presenting knowledge-related challenges.
Project Team
Hong Gao
Researcher
Haochun Huai
Researcher
Sena Yildiz-Degirmenci
Researcher
Maria Bannert
Researcher
Enkelejda Kasneci
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Hong Gao, Haochun Huai, Sena Yildiz-Degirmenci, Maria Bannert, Enkelejda Kasneci
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