一种基于时空知识图谱的意图识别置信度评估方法  

Intention Recognition Confidence Evaluation Based on a Spatiotemporal Knowledge Graph

作　　者：刘洋 刘颢[1,2] 曲腾腾 陈炜 LIU Yang;LIU Hao;QU Tengteng;CHEN Wei(Wuhan Institute of Digital Engineering,Wuhan 430205,China;School of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;School of Engineering,Peking University,Beijing 100871,China)

机构地区：[1]武汉数字工程研究所,武汉430205 [2]上海交通大学电子信息与电气工程学院,上海200240 [3]北京大学工学院,北京100871

出　　处：《指挥与控制学报》2024年第2期154-161,共8页Journal of Command and Control

基　　金：国家自然科学基金(62076249);国防科技基础加强技术领域基金(2022-JCJQ-JJ-0287);山东省自然科学基金创新发展联合基金(ZR202209130044)资助。

摘　　要：意图识别算法存在识别结果缺少评估、领域知识不完备等问题,迫切需要研究置信度评估方法。利用时空知识图谱统一表示包含时空信息的实体,将作战目标及其关系抽象表示为时空知识三元组。利用典型对抗场景数据训练神经网络,计算并融合实体和知识图谱两个层面的置信度,得到最终置信度评估结果。仿真结果表明,利用时空以及目标型号信息,分析作战目标存在某种作战意图的可能性,能有效评估意图识别结果的置信度,对于意图识别系统的真正“落地”具有重大意义。Current intent recognition algorithm has problems such as lack of evaluation of recognition results and incomplete domain knowledge,and there is an urgent need to study confidence level evaluation methods.The algorithm uses the spatiotemporal knowledge graph to uniformly represent entities containing spatiotemporal information,and abstractly represents the combat targets and their relationships as a triplet of spatiotemporal knowledge.Neural networks are trained with typical confrontation scene data,and the confidence levels of the entity and the knowledge graph are calculated and fused to obtain the final confidence level evaluation results.Simulation test results show that the method uses space-time and target model information to analyze the possibility that there is a certain combat intent in the combat targets,the confidence level of the intent recognition results is effectively evaluated,it is of great significant for the real"landing"of the intent recognition system.

关 键 词：知识图谱 时空编码 意图识别 置信度评估 

分 类 号：E91[军事]