毁伤效能大数据生态系统与知识推理模型设计

Design on damage efficiency big data ecosystem and knowledge reasoning model

针对弹药毁伤效能评估需求,该文对毁伤效能大数据资源体系建设的关键环节和相关流程进行了研究,提出了毁伤效能数据生态的概念,设计了毁伤效能大数据生态系统架构。提出结合一阶逻辑推理与知识嵌入表示的知识推理模型,解决知识库构建问题。基于该模型设计了毁伤规则挖掘算法。在FB15k和FB15k-237数据集上的实验结果表明,该文的知识推理模型的平均倒数排名(MRR)分数分别以2.0和1.5高于图查询嵌入(GQE)、Beta嵌入(BetaE)、Q2B模型中表现最好的模型。

Aiming at the demand of ammunition damage efficiency assessment,the key links and related processes of the construction of damage efficiency big data resource systems are studied here.The concept of damage efficiency data ecology is proposed,and an overall structure of the damage efficiency data ecosystem is designed.A knowledge reasoning model combining first-order logical reasoning and knowledge embedded representation is proposed to solve the damage knowledge base construction problem.A damage rule mining algorithm is designed based on the model.On the FB15k and FB15k-237 datasets,the average reciprocal ranking(MRR)scores of this knowledge reasoning model are 2.0 and 1.5 higher than those of the best performing models in the graph query embedding(GQE),Beta embedding(BetaE)and query2box(Q2B)models respectively.