普适计算中上下文依赖的主动任务发现

任务计算是一种针对用户如何与设备和服务进行交互的新模式,用户可以只关注需要完成的任务,而不必考虑如何去完成。主动任务发现是任务计算的核心,它依赖于上下文,并能自动组合相应的服务完成规定的操作。在构建任务计算模型的基础上,描述了一种基于上下文依赖的主动任务发现模式,分析了主动任务的结构模型,提出了一种发现任务并执行任务的算法,并通过一个智能教室的应用环境说明了该算法的具体应用。     

任务计算中基于语义Web的上下文感知服务选择

针对普适环境中任务计算的特点,分析了任务计算的层次结构和复杂任务的分解策略,设计了一种任务计算中服务发现的体系结构。此体系结构依赖于服务和内容的语义表示,而这种语义表示是基于本体的、共享的。它不仅可以增强精确度和调用率,而且能够实现知识共享、基于能力的搜索、自主推理以及语义匹配等功能。据此,提出了一种基于语义Web的上下文感知动态服务选择机制,这种机制可以根据待匹配的服务的动态上下文属性来对其进行过滤和排序,从而进一步优化了发现的过程、节约了用户的时间和精力。此外,该选择机制可用于发现有用的静态或动态上下文信息,从而为用户提供最合适、最相关的服务。     

URECA: Efficient Resource Location Middleware for Ubiquitous Environment

We describe an effiective resource location framework for ubiquitous computing environments populated by a diverse set of networks,devices,services and computational entities. Our framework provides context adaptation with the aid of a middleware service to improve the quality of resource location. A resource location protocol suitable to each type of network locates resource effiectively by means of dynamic reconfiguration to the current context. Our framework is also refined by support for interoperability between different types of resource location protocols occurring across a hybrid ubiquitous network. These characteristics also reduce the control overhead for resource location,saving resource,decreasing latency and permitting a considerable degree of scalability.     

Context-aware location discovery exchange in a ubiquitous environment

It becomes challenging in order to represent,discover and exchange location information in a ubiquitous environment due to dynamic movement and interaction between mobile nodes inside.In this paper,a new method is presented in order to make location information context-aware so that organizing the format of location information and maintaining the communication between direct connected nodes in a ubiquitous environment is enabled.The structure of a contextual location information repository and a context information communication protocol is manipulated to implement the proposed features.According to the simulation results in network simulator version2(NS2),the new method has depicted good discovery success and consumed efficient service discovery bandwidth.Other network traffic,i.e.transmission control protocol,(TCP) has been simulated in the scenarios but the new location-aware method has shown its robustness with continuous context discovery process.     

Context-aware BPEL workflow system using aspect-oriented programming

Business process execution language(BPEL)is a most recognized standard workflow language.However,it is difficult to be used in the ubiquitous system computing environment because it is difficult to describe the context information in the selection of the flow through the branch.To solve this problem,we propose a new BPEL workflow system with context-awareness by using aspect-oriented programming(AOP).This system is composed of a BPEL system module and a weaving module using AOP for context-aware.The BPEL system module generates a BPEL workflow program.And the weaving module converts a context-aware mark-up language(CAML)document to the aspect-oriented program that is applied to context-aware code without modification of the existing BPEL document.We also define a new document form that is called CAML,which provides a context-aware that is not available in BPEL.The system can generate a context-aware workflow program.It is developed in a way that inserts context information using AOP to provide context-aware services.     

基于谓词检测方法的上下文感知案例研究

当前,开发灵活的适应环境变化的上下文感知应用较为复杂。在上下文感知应用开发过程中,很多未知情况无法事先被充分考虑到,实际的开发又需要处理很多意外情形,因此有效地表示和处理上下文尤为重要。谓词检测作为实现上下文感知的重要方法之一,能够有效对上下文信息进行建模,但在实际应用中对于真实物理设备的支持如何在很大程度上仍然是未知的。为了应对以上问题,通过构建一个简单的物理场景,将谓词检测方法应用于真实的机器小车上,支持其在一个既定轨道上完成运行任务。在该场景下对原始的上下文环境进行建模,将环境特性逐步规约成特定的快照谓词和序列谓词,并在案例研究中,对规约化的谓词进行检测,将谓词检测应用于真实的机器小车实验上。实验分析的结果表明,谓词检测的方法能有效检测小车的上下文特性并成功支撑小车完成既定的行走目标。     

一种多层分布式情境装配模型

提出一种以Agent技术为基础的多层分布式的虚拟场景模型。先从装配体的视角构建装配信息模型,根据刻画特征对象的各项参数定义装配知识层;装配决策层将零件视为Agent,并定义Agent间传输状态参数的消息模板,同时根据虚拟装配任务的过程特点,建立决策规则;最后,定义用户控制层的运动控制算法,实现对用户的运动引导。实验表明该场景模型可在虚拟场景中有效呈现装配任务,并根据上下文情境准确理解用户意图。     

Adapting the User Context in Realtime: Tailoring Online Machine Learning Algorithms to Ambient Computing

Ambient systems weave computing and communication aspects into everyday life. To provide self-adaptive services, it is necessary to acquire context information using sensors and to leverage the collected information for reasoning and classification of situations. To enable self-learning systems, we propose to depart from static rule-based decisions and first-order logic to define situations from basic context, but to build on machine-learning techniques. However, existing learning algorithms show substantial weaknesses if applied in highly dynamic environments, where we expect accurate decisions in realtime while the user is in-the-loop to give feedback to the system’s recommendations. To address ambient and pervasive computing environments, we propose the FLORA—multiple classification (FLORA-MC) online learning algorithm. In particular, we enhance the FLORA algorithm to allow for (1) multiple classification and (2) numerical input values, while improving its concept drift handling capabilities; thus, making it an excellent choice for use in the area of ambient computing. The multiple classification allows context-aware systems to differentiate between multiple categories instead of taking binary decisions. Support for numerical input values enables the processing of arbitrary sensor inputs beyond nominal data. To provide the capability of concept drift handling, we propose the use of an advanced window adjustment heuristic, which allows FLORA-MC to continuously adapt to the user’s behavior, even if her/his preferences change abruptly over time. In combination with the inherent characteristics of online learning algorithms, our scheme is very well suited for realtime application in the area of ambient and pervasive computing. We describe the design and implementation of FLORA-MC and evaluate its performance vs. state-of-the-art learning algorithms. We are able to show the superior performance of our algorithm with respect to reaction time and concept drift handling, while maintaining an excellent accuracy. Our implementation is available to the research community as a WEKA module.     

上下文感知的树递归神经网络下隐式情感分析

针对现有的序列化模型对中文隐式情感分析中特征信息提取不准确以及对篇章级的文本信息提取存在的梯度爆炸或者梯度消失的问题,提出了双向长短时神经网络和上下文感知的树形递归神经网络(context-aware tree recurrent neutral network,CA-TRNN)的并行混合模型.该模型分别利用双向循环长...     

构造具备自适应能力的软件

随着应用场景的变迁和自身复杂性的增长,软件需要具备主动适应环境变化的能力,能够依据环境变化动态调整其行为.软件自适应的实现技术跨越了软件监控、上下文敏感计算、决策和控制理论、软件演化和维护等多个学科分支,如何系统化地构造此类软件,是软件工程领域所面临的巨大挑战.从自适应软件构造与实现这一角度出发,以“感知-决策-执行”软件自适应基本周期为主线,对已有的研究和实践进行综述.给出软件自适应的概念内涵,概述软件自适应活动在感知、决策、执行各环节上的特征分类,阐述面向自适应软件构造、关注程度较高的一系列使能技术,进而在分析典型研究项目现状的基础上,给出自适应软件构造领域的未来主要研究趋势.