Ontology based context aware situation tracking
Ontology based context aware situation tracking
复制标题
基于本体的情境感知态势跟踪
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Komal Sinha
中科院分区：
文献类型：
--
作者：
Anupama Mallik;Anurag Tripathi;R. Kumar;S. Chaudhury;Komal Sinha
无处不在的智能设备使得以无缝方式向人们提供智能服务成为可能。上下文感知有助于了解当前的事件状态或系统当前所处的情况。这种理解有助于物联网应用根据一段时间内不断变化的情况提供更相关、更智能的服务。在这篇文章中，我们提出了一种新的利用本体的情境感知情境跟踪框架。本体论代表了一个动态世界的概念模型，其中的情况在不断变化的上下文中随着时间的推移而演变。本体提供了基于输入的上下文数据以及先前情景的过去信息来推断情景的推理框架。基于当前情况和传入的上下文数据，可以怀着某种信念预测未来的情况。从传感器设备和外部输入获取上下文数据。对于每一种识别的情况，系统都会建议一些操作来提供情景感知服务。我们使用多媒体Web本体语言(MOWL)来表示本体。MOWL提出了一种与上下文观察相关的不确定性推理的概率框架。它利用动态贝叶斯网络来预测和跟踪动态变化的情况。我们举例说明了此框架在Smart Mirror用例中的使用。
Ubiquitous intelligent devices have enabled provision of smart services to people in seamless way. Context-awareness helps understand current state-of-affairs or the situation in which presently the system is. This understanding helps the IoT application provide more relevant and smarter services based on situations that change over a period of time. In this paper, we propose a novel context-aware situation-tracking framework that makes use of an ontology. The ontology represents the conceptual model of a dynamic world, where situations evolve over time in changing contexts. The ontology provides the reasoning framework to infer about a situation based on the input context data as well as the past information of earlier situations. Future situations can be predicted with some belief based on current situation and incoming context data. The context data is acquired from sensor devices and external inputs. For every recognized situation, system recommends some actions to provide context-aware service. We use Multimedia Web Ontology Language (MOWL) to represents the ontology. MOWL proposes a probabilistic framework for reasoning with uncertainties linked with observation of context. It makes use of Dynamic Bayesian networks to predict and track the dynamically changing situations. We illustrate use of this framework for Smart Mirror use case.