Sequence Generation Model Integrating Domain Ontology for Mathematical question tagging

In online learning systems, tagging knowledge points for questions is a fundamental task. Automatic technology uses intelligent algorithms to automatically tag reduce manpower and time costs. However, the current point cannot satisfy situation that mathematics often involve variable number of points, lacks consideration characteristics field, ignores internal connection between points. To address above issues, we propose Sequence Generation Model Integrating Domain Ontology Mathematical question (SOMPT). SOMPT performs data augmentation text then obtains intermediate based on domain ontology replacement facilitate deep model understand mathematical text. able obtain dynamic word vector embedding optimize textual representation math questions. What’s more, our can capture relationship tags generate more accurately in way sequence generation. The comparative experimental results show proposed has an excellent ability Moreover, generation module be applied other multi-label classification tasks par with state-of-the-art performance models.