Micro Concept Mapping AI for Automatic Syllabus Structuring

Pratiksha Ramprasad Digule

Abstract: Digital resources are growing tremendously. As a result, creating a curriculum is more complicated than it is with traditional methods that use a lot of manual expertise in developing curriculum structures; this typically causes two issues: inconsistent sequencing of concepts and lack of appropriate knowledge coverage. Over the past few years, Artificial Intelligence has emerged as a possible solution for automating curriculum development and organizing knowledge. Most existing implementations focus on extracting concepts at higher levels (i.e. the macro function) as opposed to low-level or micro conceptual extraction that builds the complete knowledge structure. The aim of this study is to introduce a new framework called MCMAI (Micro Concept Mapping Artificial Intelligence), which will automatically extract micro-level concepts from educational resources and arrange and present them in a structured way as syllabus modules, utilizing natural language processing (NLP), semantic similarity analysis and graph knowledge representation to produce a hierarchy of concepts relative to one another; thus creating a logically sequenced syllabus. The results of the experimental analysis indicate that the MCMAI framework will create a higher rate of concept coverage and improve the accuracy of prerequisites compared to traditional curriculum development/structure methods. In addition, the MCMAI framework can be used by educators, curriculum developer and adaptive learning environments to generate structured syllabi from large amounts of educational data automatically.

Keywords: Artificial Intelligence, Concept Mapping, Knowledge Graph, Natural Language Processing, Curriculum Development, Automated Syllabi Development