Landrón Rivera, Brian A.

Publication Search Results

Now showing 1 - 1 of 1

Restricted
Text classification of student predicate use for conceptual change assessment
(2016-12) Landrón Rivera, Brian A.; Santiago Román, Aidsa I.; College of Engineering; Vega Riveros, José F.; Santiago Santiago, Nayda G.; Department of Electrical and Computer Engineering; Alers Valentín, Hilton
In the field of educational engineering concepts have been placed into ontological categories that reflect the nature of those concepts. Furthermore learner misconceptions occur when students assign a concept to an incorrect ontological category within their individual mental models. Predicate tests used to estimate the categorization of student conceptions have proven to be successful for conceptual change assessment but they have not been automated with the use of modern computing systems. The main goal of this research is to show how predicate test automation is possible by applying knowledge discovery in databases theory to a previously annotated dataset to achieve. The secondary goal is to find which data mining techniques can be used to extract a feature set that yields high quality text classification results. This thesis documents how the predicate test can be automated with knowledge discovery in databases techniques using data from engineering students enrolled in a U.S. midwestern public institution.