Context-Oriented Adaptation Design for Web Information

• Main Authors: Wei-Yang Shao, 邵維揚
• Other Authors: King-Chu Hung
• Format: Others
• Language: en_US
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/20383947305647908756

博士 === 國立高雄第一科技大學 === 工程科技研究所 === 94 === Abstract With the convenience, visiting website via mobile device is an increasing demand. Due to the capabilities discrepancy between mobile device and desktop, content presentation problem arose when existed content can not fit into the screen of mobile device. This content presentation problem may impact on people’s reading and comprehending ability. Content adaptation has shown significant benefits by enhancing Web accessibility to meet users’ requirements in heterogeneous environments. However, existing content adaptation servers do not supply mobile users with this facility due to the discrepancies of screen and user context. To provide mobile users with a friendly and comprehensible way in reading HTML contents, we present a new middleware called Context-oriented Content Adaptation Framework (CCAF). In CCAF, rule-based adaptation strategy is used to customize content presentation. The adaptation strategies used in adaptation process are generated according to client’s features. We refer clients’ features (or contexts in short) to the information that may iv affect their perceptions of browsing Web contents, such as clients’ preferences and sensations, situated places, as well as devices and network connectivity. An adaptation strategy is composed of selected adaptation rules based on client’s context. We used Java Expert System Shell (JESS) to implement adaptation strategies for guiding the transformation process. When a user’s context changed, new adaptation strategy will be generated automatically by rearranging adaptation rules. For mobile screen display, content is decomposed into object level. Original objects and transcoded objects are stored in middleware and managed by a three-layered Content Structure Model, which facilitates resource participation. Moreover, a dynamic template composition method is also proposed to prevent from the necessary of complicated stylesheet management. The experiment results indicate our Context-oriented Content Adaptation Framework can successfully adapt Web contents into other versions satisfying mobile users’ contexts and obtain superior performance.