Framework for experiential transcoding of web pages with scanpath trend analysis

Abstract

Web pages consist of different visual elements, such as menus, sidebar and highlighted textual content. Although previous research shows that such elements help sighted users to complete their tasks, they can easily be distractive to users who access pages in audio. In particular, for visually-disabled users who use screen readers, such elements can mean that they need to listen to clutter to reach the desired content. In our previous work, we show that experiential transcoding can address this problem where web pages are transcoded by considering user experience. Trending elements and their order of visit are first identified by analyzing the eye movements of sighted people with Scanpath Trend Analysis (STA) and then the web pages can be transcoded by removing non-trending elements and re-ordering the trending ones based on the order they are visited. In this paper, we present our ongoing work on the development of a framework to automate and support experiential transcoding.