Cell phones and other mobile devices are characterized by limited resources, such as small screen size, limited network bandwidth, and short battery life. Without customization of content, users have to endure a significant degradation in their experience when they browse the Web on their mobile devices. Today, Web content is customized for mobile devices using manual techniques, such as WAP, that require content creators to maintain multiple versions to support a plethora of devices. As a result, the high cost associated with hand-tailoring content has limited the deployment of these techniques to a small set of Web sites and a few popular devices.
A promising alternative is to deploy an automatic adaptation service that transforms content for a wide range of devices. The main challenge for automatic content customization lies in the design of effective customization policies. This is a complex problem because optimal customization often depends on the usage semantics of the content and the user's context. For example, when adapting an image-rich web page, it may be appropriate to load high-fidelity versions of images that are central to the tasks, while loading other images at lower fidelity to save resources.
This project researches Usage awaRe Automatic Content Adaptation (URICA), a novel technique for automatic content customization that learns how to adapt content from feedback provided by users. When serving content to a mobile user, URICA first makes an initial prediction on how to customize the content. Next, URICA allows users who are unsatisfied with the system’s adaptation decision to take control of the adaptation process and make changes until the content is suitably adapted for their purposes. For example, a user may choose to remove a toolbar to improve readability, or ask the system to improve the resolution of a specific image. The successful adaptation is recorded and used in making future adaptation decisions for other users.