Learning management system based on Mixed Diagnostic Tests and Semantic Web technology

The paper describes an approach to the construction of learning management systems based on combined usage of Mixed Diagnostic Tests (MDT) in order to maintain education quality control and Semantic Web technology stack for purposes of storing learning courses' data and structures containing MDT. Mixed Diagnostic Tests are a compromise between unconditional and conditional components. Most of learning management systems being used in educational practice only utilize a binary approach to estimating quality of education: "test passed" or "test not passed". Mixed Diagnostic Tests utilize and fuzzy logic apparatus, allowing to differentiate such estimations. Central element of Semantic Web technology are semantic networks (ontologies) that are suited well to be used as a knowledge base for a system including MDT. Ontologies allow to describe both elements of an educational course (course, theme, lecture) and material being studied during the course (events, dates, terms, theorems etc.) in both full and flexible way. A set of MDT can be associated with every element of ontology describing teaching material. These MDT can be used to perform a check of mastering quality of a teaching course educational elements. Three basic ontologies, CourseLevel, KnowledgeLevel and SystemLevel, are suggested to be used in an approach being described. A CourseLevel ontology describes elements of a learning course and structures containing MDT. KnowledgeLevel ontology describes possible "knowledge elements" that will be studied during the course. SystemLevel ontology is intended for structuring CourseLevel ontologies in a framework of a single course (for example, for determination of themes' order). That ontology also stores descriptions of different resources connected with course' "knowledge elements": audio and video resources, images, books and articles. These ontologies can be supplemented with additional "child" ontologies describing in details a knowledge domain being discussed during the course, if needed. A creation of a web service for educational courses planning seems to be the simplest and most effective way of implementing such a learning management system. That web service should be used by lecturers in the network of a faculty or the university. The paper describes general architecture of a learning management system based upon the approach being suggested. It includes a database containing system users' data, knowledge base containing the above-mentioned ontologies, and a web server providing course models creation and editing. The system based upon the approach being described will possess both flexibility of description of learning courses and their elements and possibility of making a differentiate estimation of education quality for these courses.

Keywords

Authors

References