共查询到20条相似文献,搜索用时 125 毫秒
1.
2.
抽取作战文书中的军事命名实体关系,是实现作战文书语义理解的一种有效方法。在分析作战文书中军事命名实体词语规则的基础上,提出了一种结合词语规则和 SVM 模型的军事命名实体关系抽取方法。首先,使用词语规则整合作战文书中连续出现的军事命名实体并抽取其关系,使其更加适合 SVM 模型。然后,使用 SVM 模型对传统规则模板难以使用的词窗、词性和距离等特征进行建模,抽取军事命名实体关系。实验结果表明,优先利用词语规则能充分提高 SVM 模型抽取军事命名实体关系的效果,与单纯使用 SVM 模型相比,准确率和召回率分别提高了8.73%和41.71%。 相似文献
3.
针对传统军事命名实体识别方法存在人工构建特征复杂和军事文本分词不准确等问题,提出了一种基于深度学习的军事命名实体识别方法。结合双向长短时记忆(Bi-directional Long Short-Term Memory,Bi-LSTM)神经网络对较长句子上下文的记忆能力、字向量(character embedding)对汉字语义的表示能力和条件随机场(Conditional Random Field,CRF)对标注规则的学习能力,构建了character+Bi-LSTM+CRF实体识别模型。为验证方法的有效性,在军事想定语料集上进行了实验,结果表明:该方法比传统方法识别效果好,识别准确率、召回率和F值均大幅提升。 相似文献
4.
军事领域文本中存在大量军事实体信息,准确识别这些信息是军事文本信息提取和构建军事知识图谱的基础性任务。首先,提出了一种基于RoBERTa预训练模型、跨度和对抗训练的标签指针网络的融合深度模型(RoBERTa-Span-Attack),用于中文军事命名实体识别;然后,采用了一种基于Span的标签指针网络,同时完成实体的起止位置和类别的识别任务;最后,在模型训练过程中加入对抗训练策略,通过添加一些扰动来生成对抗样本进行训练。在军事领域数据集上的实验结果表明:所提出的军事领域命名实体识别模型相较于BERT-CRF、BERT-Softmax和BERT-Span,在识别准确度上具有更优的效果。 相似文献
5.
6.
将卷积结构引入循环神经网络,从而构建卷积循环神经网络。以此为基础,研究构建了面向中文分词与实体识别联合学习的序列标注模型。该模型依托卷积循环神经网络构建特征编码层,实现中文字序列局部空间特征和长距离时序依赖特征的联合提取;依托改进的循环神经网络构建标签解码层,实现标签序列长距离时序依赖的有效建模;依托统一的分词与实体识别序列标注模式实现分词信息与实体信息的联合学习,避免传统流水线法的误差传播问题。在人民日报语料和微软标注语料上的实验结果显示,该框架较传统统计模型和神经网络模型有显著的性能提升,尤其是在识别字数较多的命名实体时,其效果明显优于其他方法。 相似文献
7.
8.
9.
由于军事系统中存在不同的体制编制和多变的网络结构,并且军事单元能力发挥不稳定,信息延时不确定,因此为完成特定军事任务,系统中会存在诸多OODA过程,以至于难以评估军事系统的效能.针对目前效能评估方法人为定性评估成分较大,或者评估过程忽视军事实体间的相关联系而仅在统计意义上进行评估,将军事系统的不确定性纳入评估体系,基于统计关系学习框架Markov logic以OODA军事策略为基础对不同军事规则进行一阶逻辑建模,并以当前网络态势为先验条件,应用Alchemy工具对表示最终作战效能的原子谓词进行条件概率推理,求得该网络作战效能期望,突出了模型对不确定信息的处理能力.最后研究了军事单元和军事体制对系统效能的影响,实验结果验证了本模型的有效性,表明其具有一定的实践和理论意义. 相似文献
10.
11.
12.
新军事变革和多样化军事任务为当前指挥控制系统带来了前所未有的挑战,研究表明面向服务协作(Service-oriented Collaboration,SOC)的应用可以有效增强信息系统的实时信息共享和协同能力,但现有平台存在自适应性和扩展性不强等问题。文章通过引入灵活对象技术,设计提出了一种基于灵活对象的服务协作应用框架,并基于指挥控制系统建立了原型系统。通过应用分析表明,本文所提出的服务协作应用框架具有很好的自适应性和扩展性,有利于信息化条件下的指挥控制系统快速构建的要求。 相似文献
13.
The modern Mission Command philosophy in military operations originated with the German concept of Auftragstaktik (mission tactics). It emphasizes in general terms the exercise of disciplined initiatives by subordinates during the execution of mission-type orders in the course of military operations that are conducted within the overall intent of a commander. The present paper contains an overview of the historical evolution of the mission command concept and its application. The paper compares the mission command concept with the detailed or linear method of command. This paper focuses on the relevance of the mission command concept for the Hellenic Land Forces (HLF) in Greece. The paper provides a historical account on when and where mission command was successfully exercised by the Hellenic Armed Forces in the modern military history of Greece. The authors support the premise that the mission command concept is essential for safeguarding Greek national security in the current environment of evolving symmetric and asymmetric threats, and in meeting operational challenges in a diverse and difficult terrain. The paper examines the cultural reasons that have impeded the adoption of the mission command concept within the HLF and recommends a three-phase model for its permanent implementation. 相似文献
14.
15.
Theory on the use of information technology in military operations assumes that bringing together units in an information network helps units to work together. Decentralized command systems such as mission command have been proposed for these networks, so that units can adapt to changes in their turbulent working environments. Others have proposed centralized command systems that permit higher organizational levels to closely direct military operations. This article uses Perrow’s (1984, 1999) Normal Accidents Theory to propose that increasing interdependencies between units in information networks places incompatible demands on the design of networked military operations. It is concluded that networked military operations require decentralized command approaches, but only under the condition that interdependencies between modules of networked units are weak rather than tight. This precondition is essential for retaining control over networked military operations. 相似文献
16.
17.
《防务技术》2022,18(11):2083-2096
Ground military target recognition plays a crucial role in unmanned equipment and grasping the battlefield dynamics for military applications, but is disturbed by low-resolution and noisy-representation. In this paper, a recognition method, involving a novel visual attention mechanism-based Gabor region proposal sub-network (Gabor RPN) and improved refinement generative adversarial sub-network (GAN), is proposed. Novel central–peripheral rivalry 3D color Gabor filters are proposed to simulate retinal structures and taken as feature extraction convolutional kernels in low-level layer to improve the recognition accuracy and framework training efficiency in Gabor RPN. Improved refinement GAN is used to solve the problem of blurry target classification, involving a generator to directly generate large high-resolution images from small blurry ones and a discriminator to distinguish not only real images vs. fake images but also the class of targets. A special recognition dataset for ground military target, named Ground Military Target Dataset (GMTD), is constructed. Experiments performed on the GMTD dataset effectively demonstrate that our method can achieve better energy-saving and recognition results when low-resolution and noisy-representation targets are involved, thus ensuring this algorithm a good engineering application prospect. 相似文献
18.
19.
20.
LED显示屏是现代社会发布信息的一种重要方式,通过LED显示屏可以播放视频、文字等不同形式的信息,为人们传达最直观的视觉信息。随着科学技术的不断发展,LED显示屏的应用也日渐广泛,尤其是军事上的应用,比如基于LED显示屏的各种级别指挥控制终端的设计和运用,具有更为深入的研究价值。而ARM处理器是实现LED显示屏控制系统有效运行的基础,主要针对基于ARM的LED显示屏指挥控制终端系统的设计与实现的相关问题进行了探讨。 相似文献