面向移动Web的情境化网构软件开发技术研究

In our undergoing National Science Foundation project, we focus on the development of situational Internetware. Taking web applications as our target, we have proposed a systematic approach for situational web applications development for adapting situation changes. The approach is user-driven, and leverages the collaboration between client browsers and server-side supports. Results have been published on several high-quality journals and conferences, cited and praised by some related researchers, and obtained several academic awards. .In mobile Internet environment, developing situational web applications becomes challenging, due to the diversity of mobile devices and their computation resources, requirements for heterogeneous applications collaboration, personalized user preferences and behaviors, etc. It brings new opportunities of Internetware research. We try to continue and extend our approach for situational web application development in this continuous project. We will leverage the mobile cloud computing paradigm, make full use of mobile device capabilities, and the synergy of mobile client and cloud. Key problems include: (1)the situation representation model of mobile web applications; (2)the component model and programming framework, and(3) the client-cloud synergy development supports.

项目组在青年科学基金项目中，以情境化Web应用为研究对象，系统化地研究了以用户偏好导向、基于客户端和服务器端交互协同的网构软件开发方法并研制原型，使得Web应用灵活、快速适应情境变化，研究成果得到了国内外同行的认可和跟进。.随着互联网迅速发展和移动终端普及，Web应用面临着适应终端设备多样化、计算资源差异化、异构应用协同化、使用方式个性化等情境性问题，也为网构软件开发带来了新的机遇和挑战。基于青年基金成果，本项目将以“充分发挥移动端能力特征并支持移动端和服务器端的协同融合”为切入点，重点突破移动Web应用情境性及其描述规约、情境化Web应用构件模型和移动客户端-服务器融合化开发支持机制三个关键问题，从原理、方法和实现三个层面，系统化地探讨面向移动Web的情境化网构软件开发方法与技术。

项目组将研究对象从互联网环境下运行于单机的Web应用扩展到移动互联网环境下运行于多种异构设备上的移动Web应用，以“充分考虑端设备能力特征并支持云（服务社区）和端（设备）协同融合”为切入点，系统地研究情境化移动 Web 应用开发方法与技术。主要研究包括：1）针对移动环境下复杂的情境，建立用户数据驱动的移动 Web 应用情境模型。提出一种基于普适语言的情境感知方法，多角度理解用户情境。运用软件静态分析技术，研究后台唤醒情境下移动应用性能特征。2）提出一种面向情境的SMVC构件模型。结合情境信息，提出基于用户行为的移动Web应用质量评价方法。运用信息抽象技术与图论算法，实现应用服务的发现和组装。3）提出一种基于云-端协同数据缓存的移动Web应用性能优化机制。围绕上述内容，项目组共发表国际顶级期刊、会议论文17篇，申请专利3项。