Csurf: A context-driven non-visual web-browser

Web sites are designed for graphical mode of interaction. Sighted users can "cut to the chase" and quickly identify relevant information in Web pages. On the contrary, individuals with visual disabilities have to use screen-readers tobrowse the Web. As screen-readers process pages sequentially and read through everything, Web browsing can become strenuous and time-consuming. Although, the use ofshortcuts and searching offers some improvements, the problem still remains. In this paper, we address the problemof information overload in non-visual Web access using thenotion of context. Our prototype system, CSurf, embodyingour approach, provides the usual features of a screen-reader.However, when a user follows a link, CSurf captures thecontext of the link using a simple topic-boundary detectiontechnique, and uses it to identify relevant information onthe next page with the help of a Support Vector Machine, astatistical machine-learning model. Then, CSurf reads the Web page starting from the most relevant section, identifiedby the model. We conducted a series experiments to evaluate the performance of CSurf against the state-of-the-artscreen-reader, JAWS. Our results show that the use of context can potentially save browsing time and substantiallyimprove browsing experience of visually disabled people.