泛在学习及其关键技术

泛在学习为教育改革提供了新的视角.依据对泛在学习主要特征的分析,提出了泛在学习环境模型,并讨论了支持泛在学习的关键技术.     

Context-Aware Using Carrier Phase for Adaptive MEMS IMU/GNSS Filtering in Deep Urban Navigation

Context-aware is becoming standard on the most mobile navigation devices. The performance of MEMS IMU/GNSS gains significant benefits from context information in terms of improvement of filter’s adaptive capability. A context-aware algorithm using differential carrier phase was proposed to recognize a mobile MEMS IMU/GNSS equipped vehicle’s stationary, slow moving or fast moving status. The corresponding context error in awarding was analyzed and consequently conducted two fading factors based on the analysis. The factors were applied in the system’s adaptive filter with targeting applications in deep urban where severe multipath presents. The dense urban field test shows that the false alarm of proposed context-aware algorithm is less than 5％ and the adaptive filtering can achieve around 15％ improvement in terms of 1σ in two-dimension position accuracy.     

位置感知计算的概论、关键问题和技术探讨

自从Mark Weiser在1991年提出普适计算的思想以来,计算模式经历主机计算、桌面计算的阶段后正朝普适计算方向蓬勃发展.上下文感知是普适计算的最重要的特征,同时也是主要的研究领域.自从二十世纪70年代美国军方推出军民两用的室外定位系统GPS,以及1989年Roy Want等开发出支持大规模使用、价格低廉的室内位置传感系统Active Badge以来,位置感知由于其高鲁棒性和确定性、技术的可获得性而成为当前上下文感知计算的一个真正的研究热点.本文从普适计算和智能空间的特征和要求出发,阐述了位置感知计算的主要概念、思想,并根据当前最新的研究状况,解释了位置感知计算面临的主要问题和相应的技术方法,最后叙述了作为本课题组研究的试验系统--智能教室的技术路线.     

基于Spark的实时情境推荐系统关键技术研究

随着移动互联网的飞速发展,人们面临的信息过载的问题日益严重,大数据场景下对用户的推荐面临着巨大困难。为了解决推荐时效性、准确度、大数据量,提出了一种基于Spark的实时情境推荐算法。该算法在协同过滤的基础上融合了情境过滤,以Kafka作为实时消息收发器,以Spark Streaming来处理实时流数据,增强了算法的准确性和时效性。实验证明,该算法和传统协同过滤算法相比,准确率和时效性更高,且在大数据场景下更有优势。     

Location Based and Customized Voice Information Service for Mobile Community

Mobile commerce has been increasingly recognized as one of the most important prosperous areas for deploying information technology. This paper aims to advance the value of the information by providing a novel voice-information sharing mechanism that is a combination of a location-based information service (known as a Killer App) and a virtual community that consequently becomes a WCP (Wireless Content Provider). This community embodies context aware intelligence by analyzing the context-sensitive behavior of the community members and enabling proactive and precise context sensitive voice-information sharing; furthermore. This voice-information sharing mechanism is comprised of an IVR system, a location service, EPN (Euclidean distance with Positive and Negative Strength) Clustering, Naïve Bayesian Prediction, and a set of metrics for monitoring the progression of the community. The primitive results show that our mechanism satisfactorily reaches the goal of proactive precise sharing of voice information between community members.     

Context-Aware信息检索技术与模型

Context-aware信息检索技术是一种具有自适应性和自组织性的智能信息检索技术,而且是富有挑战性的研究方向。文章首先叙述了与Context-aware相关的基本概念及实现的基本原理,接着介绍了基于不同Context信息的检索模型,并且讨论了各种模型的实现技术及适用范围。     

普适计算环境中上下文模型的建立

普适计算作为第三代计算方式已经成为科学研究的前沿,上下文作为普适计算中最为重要研究方向之一,对普适计算的研究和发展有着大力地推动作用.就建立普适计算环境上下文模型进行了分析和讨论,对上下文的数据和通信方式也进行了分析,为上下文的实际应用提供一些理论基础准备.