Date of Award

1-1-2020

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

Department of Computer Science

Content Description

1 online resource (xiii, 132 pages) : color illustrations.

Dissertation/Thesis Chair

Shaghayegh Sahebi

Committee Members

Reza Feyzi-Behnagh, Chelmis Charalampos, Ming-Ching Chang

Keywords

Educational Data Mining, Matrix Factorization, Sequential Pattern Mining, Student Behavior, Study skills, Decision making, College students, Web-based instruction, Computer-assisted instruction

Subject Categories

Computer Sciences

Abstract

The goal of this dissertation is to examine factors such as how a student chooses to engage with the online platform and time spent on individual tasks and draw conclusions to improve the efficiency of the students and efficacy of online learning tools. Student activities and decision-making while functioning in a computer-based learning environment are utilized to guide students with effective patterns in studying. In addition to the sequence of actions, we have considered the time spent on each activity in modeling to have a more accurate representation of students' behavior in studying. Using sequential pattern mining methods, we find students' patterns of behavior in studying. By analyzing the correlation between the most frequent patterns and students' performance, we will derive effective patterns of interaction in an online learning environment. We propose a novel method using non-negative discriminative matrix factorization and pattern structures to find patterns that are common among students (learning traits) and patterns that are specific to a group of students with low or high performance (performance traits). We use these patterns to model the students' behavior and predict their performance.

Download

Share

COinS