Progressive Insights in use of Machine Learning to Support Student Engagement Diversity: The XYZ EduOwl chatbot

Arezou Shafaghat; Mohammad Jonaidi; Hoseoen Lee; Craig A Chin; Ali Keyvanfar

Download Paper | Permalink

Conference: 2024 South East Section Meeting
Location: Marietta, Georgia
Publication Date: March 10, 2024
Start Date: March 10, 2024
End Date: March 12, 2024
Tagged Topic: Diversity
Page Count: 17
DOI: 10.18260/1-2--45554
Permanent URL: https://peer.asee.org/45554
Download Count: 360

Paper Authors

biography

Arezou Shafaghat Kennesaw State University

visit author page

I am a results-oriented and motivated professor, consultant, and scientist with over ten years of international professional experience in sustainable and smart urban development.

visit author page

biography

Mohammad Jonaidi

visit author page

Dr. Jonaidi obtained his Ph.D. from Sydney University and is currently working at Kennesaw State University. During 38 years of research and industry professions, he has been involved in analysis and design of complex structural projects such as: FEA of high-rise buildings/steel structures, Floor vibration for concrete slabs/pedestrian bridges, Serviceability vibration analysis of high-rise buildings, Earthquake Engineering, Post-tensioned concrete structures, Nonlinear and buckling analysis of thin-walled cylinders, Analysis of long span spatial steel structures, Analysis of Glazing façade, Below grade shoring walls, Retrofit of concrete structures using Fiber Reinforced Polymers (FRP), and the strengthening of structures to resist progressive collapse.

visit author page

author page

Hoseoen Lee

biography

Craig A Chin Kennesaw State University

visit author page

Craig A. Chin is currently an Associate Professor in the electrical engineering department at Kennesaw State University. His research interests include applying digital signal processing and machine learning techniques to biomedical signals/images, and in engineering education.

visit author page

biography

Ali Keyvanfar Kennesaw State University

visit author page

Dr. Ali is a full-time Faculty in Construction Management at Kennesaw State University. His most recent attachments are the KSU Institutional Review Board (IRB), UN SDG faculty at CIFAL Atlanta (a former UNITAR center), ABET Program Evaluator (EAC & ETAC), Visiting Scholar at UC San Diego, Visiting Professor UTE (Ecuador), Visiting Professor at KiA (South Korea), and Managing editor of MIT sustainable city program in Universiti Teknologi Malaysia (UTM). Dr. Ali is an experienced academician, R&D project manager, journal editor, start-up investment advisor, and international consultant with close to ten years of record in sustainable construction engineering and management (by method and material). Dr. Keyvanfar is a dedicated team leader with a current focus on diversity in research.

visit author page

Download Paper | Permalink

Abstract

Personalized education emphasizes adapting educational content, engagement, and assessment wants to individual learners, departing from traditional, uniform educational models. The manuscript emphasizes the necessity of rethinking curriculum design and assessment methods to align with personalized learning. Traditional curricula and one-size-fits-all assessments may not effectively address diverse learning styles and wants. XYZ EduOwl is a tool developed to address the diverse engagement types and assessment wants of students in the modern educational landscape. It leverages machine learning techniques to identify and cater to individual styles and wants. As a work-in-progress, a simulated dataset generated using ChatGPT ADA was employed to evaluate the validation method of user perceptions of the tool through a comprehensive survey designed to gather insights into user experiences and perceptions. The manuscript explores generating normal distribution plots for each survey question, enabling a visual representation of response trends and variations. Additionally, network analysis was utilized to explore the interconnections among different aspects of user experience - educational interests (X series), engagement styles (Y series), and assessment wants (Z series). The study plans to evolve from theoretical underpinnings to practical application, incorporating extensive data and analysis from a case study conducted at Kennesaw State University. This case study will utilize a variety of courses and departments to gather substantial empirical evidence, demonstrating the tool's effectiveness in catering to individual learning styles and needs. Key findings include visual representations of user response trends through normal distribution plots and network analysis of the interconnections between educational interests, engagement styles, and assessment preferences. The manuscript highlights the crucial role of AI-driven personalization in contemporary education, supported by anonymized data and source code availability for broader academic adoption and validation.

Citation
Format

Shafaghat, A., & Jonaidi, M., & Lee, H., & Chin, C. A., & Keyvanfar, A. (2024, March), Progressive Insights in use of Machine Learning to Support Student Engagement Diversity: The XYZ EduOwl chatbot Paper presented at 2024 South East Section Meeting, Marietta, Georgia. 10.18260/1-2--45554

TY  - CPAPER
AB  - Personalized education emphasizes adapting educational content, engagement, and assessment wants to individual learners, departing from traditional, uniform educational models. The manuscript emphasizes the necessity of rethinking curriculum design and assessment methods to align with personalized learning. Traditional curricula and one-size-fits-all assessments may not effectively address diverse learning styles and wants. XYZ EduOwl is a tool developed to address the diverse engagement types and assessment wants of students in the modern educational landscape. It leverages machine learning techniques to identify and cater to individual styles and wants. As a work-in-progress, a simulated dataset generated using ChatGPT ADA was employed to evaluate the validation method of user perceptions of the tool through a comprehensive survey designed to gather insights into user experiences and perceptions. The manuscript explores generating normal distribution plots for each survey question, enabling a visual representation of response trends and variations. Additionally, network analysis was utilized to explore the interconnections among different aspects of user experience - educational interests (X series), engagement styles (Y series), and assessment wants (Z series). The study plans to evolve from theoretical underpinnings to practical application, incorporating extensive data and analysis from a case study conducted at Kennesaw State University. This case study will utilize a variety of courses and departments to gather substantial empirical evidence, demonstrating the tool&#39;s effectiveness in catering to individual learning styles and needs. Key findings include visual representations of user response trends through normal distribution plots and network analysis of the interconnections between educational interests, engagement styles, and assessment preferences. The manuscript highlights the crucial role of AI-driven personalization in contemporary education, supported by anonymized data and source code availability for broader academic adoption and validation.
AU  - Arezou Shafaghat
AU  - Mohammad Jonaidi
AU  - Hoseoen Lee
AU  - Craig A Chin
AU  - Ali Keyvanfar
CY  - Marietta, Georgia
DA  - 2024/03/10
PB  - ASEE Conferences
TI  - Progressive Insights in use of Machine Learning to Support Student Engagement Diversity: The XYZ EduOwl chatbot
UR  - https://peer.asee.org/45554
DO  - 10.18260/1-2--45554
ER  -