Entity Recognition Systems for Structuring Database in Industrial Robot Maintenance Education

The growing adoption of industrial robots in manufacturing and automation sectors has emphasized the need for effective education and training systems in maintenance practices. Traditional training methods often rely on manual documentation and unstructured learning materials, which limits knowledge transfer and hampers students' ability to quickly identify and understand key technical concepts. The problem lies in the absence of automated tools that can extract, classify, and structure domain-specific knowledge from diverse maintenance resources such as manuals, technical reports, and instructional notes. To address this gap, the paper propose the Entity Recognition for Industrial Robot Maintenance Education (ER-IRME) framework, which applies advanced natural language processing (NLP) and named entity recognition (NER) techniques to identify critical entities such as components, fault types, maintenance procedures, and safety measures. ER-IRME then organizes these entities into structured knowledge bases, enabling more efficient search, visualization, and adaptive learning. Experimental evaluation on a curated dataset of industrial robot maintenance documents demonstrated that ER-IRME achieved an F1score of 91% in entity extraction, improved knowledge retrieval efficiency by 34%, and enhanced student learning outcomes by 29% compared to traditional keyword-based approaches. These results confirm that entity recognition systems can significantly improve both the accessibility and quality of robot maintenance education. In conclusion, the ER-IRME framework provides a scalable and intelligent solution to structuring unstructured maintenance knowledge, facilitating effective education and bridging the gap between theory and practice in industrial robotics training.

Ҳали таржима қилинмаган