Knowledge Graph Construction for Foreign Military Unmanned Systems

Yilin Chen, Jingting Wang, Shutong Zhu, Yuang Gu, Haoyu Dai, Jingyi Xu, Yipeng Zhu, Tianxing Wu

Abstract: Unmanned systems have become a significant component of modern military forces and play a more and more important role in various military operations. It mainly consists of four major domains which are unmanned aerial system (UAS), unmanned ground vehicle (UGV), unmanned underwater vehicle (UUV), and unmanned surface vessel (USV). This paper focuses on the construction of a high-quality knowledge graph on foreign unmanned systems and proposes an effective method to complete the construction. The method first analyses the data provided by CCKS2022 evaluation organizers and builds a schema. Then not only data provided are used to construct the knowledge graph but also external data are crawled and extracted as triples under constraints of the schema. After that, entities are aligned and logic rules are also utilized to knowledge graph completion. Finally, the knowledge graph constructed is stored and visualized in the Neo4j database and evaluated by the question-answering tasks. This paper presents our technics for the 13th task of CCKS2022 evaluation (i.e. knowledge graph construction and evaluation on foreign military unmanned systems) and our team win the 3rd place in this task.