Machine Learning Founded Domain Vocabulary Creation for Enriched Concept Map Mining

Knowledge may be successfully organised, represented, and shown with the use of something called a concept map. A concept map is a form of knowledge representation graph that is more loosely semantic. A clear and succinct summary of the material needs to be presented via the idea map. This overview should be simple to grasp. An essential component of a concept map is its ability to accurately represent and organise ideas while simultaneously providing access to the greatest possible amount of information. In the ever-evolving realm of domain vocabulary creation, various methodologies have been posited, each with its own strengths and weaknesses. This study embarks on a comparative analysis of traditional techniques vis-à-vis a newly proposed machine learning (ML)-based approach, evaluated across three tables focusing on different dimensions: accuracy and computational performance, robustness and versatility, and user experience and efficiency. Through a comprehensive comparison, we aimed to discern the most optimized method, and our results markedly indicate the ML-based approach's superiority across most parameters. From achieving the highest precision, recall, and F1-score, to its adaptability across languages and domains, and its intuitive interface coupled with real-time operation capabilities, the proposed method stands out as a next-gen solution for domain vocabulary creation.