M.S. (Master of Science)
Department of Computer Science
Computer science; Education
A concept map can be used to represent knowledge. A causal map is a type of concept map and a graphical way that can be used to define thinking of an individual about a system by using links between nodes in a network to represent causation or influence. A causal map thus can be used to solve ill-structured problems and evaluate student's assignment. Causal maps can also be used to generate feedback for a student by comparing it with an expert map. However, exact network comparison is an NP-complete problem and is known to be graph isomorphism problem. In order to solve the problem, an online platform is created, where instead of exact graph comparison mathematical techniques such as graph edit distance, graph kernel, and graph embedding are used to measure how closely two networks match. Before performing comparison it is important to establish a common set of terms between two networks, as a student term might have a slight variations in the way they name the included concepts in causal map. This requires alignment of terms which is often contextual, the process is eased by implementing a recommendation system by leveraging alignments done by other users. Step by step feedback is generated to guide the student in developing their understanding of a system. This software can be used in Massive Open Online Course (MOOC) and classroom-based system, where assessing the learning experience and providing feedback to a large number of participants is an ongoing challenge.
Gupta, Vishrant Krishna, "Efficiently comparing mental representations : visualizing and matching causal networks" (2018). Graduate Research Theses & Dissertations. 2817.
Northern Illinois University
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.
Advisors: Reva Freedman.||Committee members: Hamed Alhoori; Nicholas Karonis.||Includes illustrations.||Includes bibliographical references.