Technical Report 2017-18-01
Internetworking and Media Communications Research Laboratories
Department of Computer Science, Kent State University


Dissertation Title: 

An Assessment of Knowledge by Pedagogical Computation on Cognitive Level Mapped Concept Graphs

Rania Anwar Aboalela

Advisor: Prof. Javed I. Khan
Department of Computer Science
Kent State University

  August 2017


This research presents a novel learning assessment model to measure the student learning in terms of learning cognitive skill levels and the current knowledge state towards achieving the level. Knowledge state refers to if a student has already learned, or ready to learn, or not ready to learn a certain concept.  The cognitive skill levels refers to levels such as if a student has acquired the state at the level of understanding, or applying, or analyzing, etc. a particular concept. The cognitive skill levels are based on Bloom’s taxonomy (Anderson, et al., 2001).  The knowledge states are defined based on the Knowledge Assessment theory proposed by (Falmagne, Cosyn, Doignon, & Thiery, 2003). The model is comprised of four constructions. First we propose a semantic/ ontological scheme called Cognitive Level Mapped Concept Graphs (CLMCG), which maps the concepts appearing in a knowledge area and their pedagogical and ontological relations as understood by the domain experts. To capture the organization of the knowledge domain, we use formal syllabus and textbooks written by the experts. The concepts are then mapped in three relational dimensions: the syllabus dimension, the ontology dimension and the cognitive skill dimension needed to perform the analysis. Currently, we extract the relationships along the first two dimensions based on the organization of the concepts as they appear in leading text book(s) and the third dimension following the classifications given by Bloom (Bloom, Engelhart, Furst, Hill, & Krathwohl, 1956). We then propose a concept based testing and evaluation scheme for testing students. We then show methods of analysis scheme based on graph traversal and logical inference that can provide the assessment of the student’s knowledge in terms to a modified Falmagne state and modified Bloom target skill levels. Finally, we also propose a new set of quantitative measures and scales that we believe can provide much insight as well as objective assessment compared to a current numerical assessment system.


Last Modified: Sep 2017.