Entity Classification for Military Knowledge Graph based on Baidu Encyclopedia Distance Learning

Entity types are a critical enabler for many NLP tasks that use KGs as a reference source. However, Classifying terminological entities without context remains an important outstanding obstacle in the field of KG completion. In this paper, we put forward a method combining distance learning and deep learning to address the classification of entity with no context. We compare the performance of our method with several text classification methods and shows our approach is empirically effective. Furthermore, the experiment result shows our approach can reduce the labeling work cost and expand the entities for further knowledge graph construction.