User-centric adaptation of Web information for small screens

A global increase in PDA and cell phone ownership and a rise in the use of wireless services lead mobile browsing to become an important means of Internet access. However, mobile devices have a small screen, which limits the usability of mobile browsing. This paper presents a novel method that automatically adapts a desktop presentation to a mobile presentation, proceeding in two steps: detecting boundaries among different information blocks (i.e., page segmentation) and then generating a user preferred adaptive layout. Distinct from other approaches, our approach analyzes both the DOM structure and the visual layout to detect closely related contents. In the process of page segmentation, our approach first divides a Web page into several common areas (such as top, bottom, and main content), and then further identifies different topics in the main content based on the visual and structural analysis. The page segmentation produces a block tree, which represents the information organization underlying a Web page. Based on the block tree, an adaptive layout, which places closely related information in proximity and minimizes scrolling, is automatically generated. All current approaches support only one fixed style (e.g., a narrow-page style) to render an adaptive layout for a diverse range of users, Websites, and devices. This "one size fits all" strategy could not offer universal usability. Our approach supports a user to flexibly specify the style of an adaptive layout according to his/her personal preferences. The theoretical foundation of such a user-centric adaptive layout is a feature space, which specifies an adaptive layout from different aspects, such as the location and presentation of navigation options. A user can visually manipulate those features to create a personalized style in a graphical user interface, and then an adaptive layout consistent with the personalized style is automatically generated. The user study based on a prototype shows the usability and efficiency of mobile browsing are significantly improved. (C) 2011 Elsevier Ltd. All rights reserved.