A Reader's Control Browser Based on Graph and HyperGraph -Guide User in Hyperspace

• Main Authors: Lung-Tsai Lin, 林龍在
• Other Authors: Hsiu-Hsen Yao
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/54165774821614659926

碩士 === 元智大學 === 實業計劃研究所工學組 === 81 === Hypertext is becoming a popular approach to many computer applications, especially those dealing with the on-line presentation of large amounts of loosely structured information such as on-line documentation. In general, hypertext uses node/link model to provide cross-reference between related information. When readers navigate in the hypertext network they often have "lost in hyperspace" and "cognitive overhead" problems, thus they cannot understand where to go, where come from, where can get the information they need. Current hypertext system pose various strategies to resolve above problems, but they often ignore the reader's mental model, so that the browser tool cannot match the requirement of different users. The thesis present the graph model to model the hypertext information space, hypergraph to model the relationships between nodes and classes(sets) and EER model to describe the relationships(semantics) between classes. As for the external presentation of information space we use the table and graph to present it. Also from cognitive science we understand that Hypertext usage depends on what mental models users have for system so we develop necessary model operators for readers' to determine the information scope they need and to establish a personal browser for further using. Finally, we choose a application area(news browsing system) to develop a prototype to verify our theory.