Automatic Domain Module Extraction using Machine Learning Approach from Electronic Textbooks

Abstract: In recent days, technology-supported learning systems (TSLSs), such as intelligent tutoring systems (ITSs), adaptive hypermedia systems (AHSs), and especially, learning management systems (LMSs) are being extensively used in many studious institutions and becoming necessary for learning. The Domain Module is measured the core of any TSLSs as it represents the data about a topic matter to be communicated to the learner. In the existing system, a DOM-Sortze is a system that uses NLP techniques, heuristic analysis, and ontologies for the semiautomatic structure of the Domain Module from electronic textbooks. But in this system, still lack in the identification of pedagogical relationships. This is needed to improve in this system. In other words, DOM-Sortze system is not able to including the new rules of the pedagogical relationships. To overcome this issue, using learning techniques to learn the new rules in the pedagogical relationships. In our proposed system, we are proposing the SVM (support vector machine) learning approach intended for learning process. Our machine learning methods are used to infer new rules in order to improve the identification of pedagogical relationships or the DRs in the electronic textbooks.

Keywords: knowledge acquisition, SVM, domain extract, ontology learning