You are here

Learning Opportunities and Challenges of Sensor-enabled Intelligent Tutoring Systems on Mobile Platforms: Benchmarking the Reliability of Mobile Sensors to Track Human Physiological Signals and Behaviors to Enhance Tablet-Based Intelligent Tutoring System

Download pdf | Full Screen View

Date Issued:
2018
Abstract/Description:
Desktop-based intelligent tutoring systems have existed for many decades, but the advancement of mobile computing technologies has sparked interest in developing mobile intelligent tutoring systems (mITS). Personalized mITS are applicable to not only stand-alone and client-server systems but also cloud systems possibly leveraging big data. Device-based sensors enable even greater personalization through capture of physiological signals during periods of student study. However, personalizing mITS to individual students faces challenges. The Achilles heel of personalization is the feasibility and reliability of these sensors to accurately capture physiological signals and behavior measures.This research reviews feasibility and benchmarks reliability of basic mobile platform sensors in various student postures. The research software and methodology are generalizable to a range of platforms and sensors. Incorporating the tile-based puzzle game 2048 as a substitute for a knowledge domain also enables a broad spectrum of test populations. Baseline sensors include the on-board camera to detect eyes/faces and the Bluetooth Empatica E4 wristband to capture heart rate, electrodermal activity (EDA), and skin temperature. The test population involved 100 collegiate students randomly assigned to one of three different ergonomic positions in a classroom: sitting at a table, standing at a counter, or reclining on a sofa. Well received by the students, EDA proved to be more reliable than heart rate or face detection in the three different ergonomic positions. Additional insights are provided on advancing learning personalization through future sensor feasibility and reliability studies.
Title: Learning Opportunities and Challenges of Sensor-enabled Intelligent Tutoring Systems on Mobile Platforms: Benchmarking the Reliability of Mobile Sensors to Track Human Physiological Signals and Behaviors to Enhance Tablet-Based Intelligent Tutoring Systems.
12 views
3 downloads
Name(s): Vazquez, Luis, Author
Proctor, Michael, Committee Chair
Jentsch, Florian, Committee Member
Gonzalez, Avelino, Committee Member
Sottilare, Robert, Committee Member
University of Central Florida, Degree Grantor
Type of Resource: text
Date Issued: 2018
Publisher: University of Central Florida
Language(s): English
Abstract/Description: Desktop-based intelligent tutoring systems have existed for many decades, but the advancement of mobile computing technologies has sparked interest in developing mobile intelligent tutoring systems (mITS). Personalized mITS are applicable to not only stand-alone and client-server systems but also cloud systems possibly leveraging big data. Device-based sensors enable even greater personalization through capture of physiological signals during periods of student study. However, personalizing mITS to individual students faces challenges. The Achilles heel of personalization is the feasibility and reliability of these sensors to accurately capture physiological signals and behavior measures.This research reviews feasibility and benchmarks reliability of basic mobile platform sensors in various student postures. The research software and methodology are generalizable to a range of platforms and sensors. Incorporating the tile-based puzzle game 2048 as a substitute for a knowledge domain also enables a broad spectrum of test populations. Baseline sensors include the on-board camera to detect eyes/faces and the Bluetooth Empatica E4 wristband to capture heart rate, electrodermal activity (EDA), and skin temperature. The test population involved 100 collegiate students randomly assigned to one of three different ergonomic positions in a classroom: sitting at a table, standing at a counter, or reclining on a sofa. Well received by the students, EDA proved to be more reliable than heart rate or face detection in the three different ergonomic positions. Additional insights are provided on advancing learning personalization through future sensor feasibility and reliability studies.
Identifier: CFE0007260 (IID), ucf:52177 (fedora)
Note(s): 2018-08-01
Ph.D.
Engineering and Computer Science, Dean's Office GRDST
Doctoral
This record was generated from author submitted information.
Subject(s): Android -- Intelligent Tutoring Systems -- mobile -- postures -- personalized learning -- puzzle game -- sensors
Persistent Link to This Record: http://purl.flvc.org/ucf/fd/CFE0007260
Restrictions on Access: campus 2019-08-15
Host Institution: UCF

In Collections