PAR: Partition-Annotate-Recommend Paradigm for Improved Accessibility in Digital Libraries

While navigation within complex information spaces is uneasy for all users, it is extremely difficult for visually impaired users who can not simply searching and browsing digital contents with a mouse. These users have to listen line by line using a screen reader program, which may be particularly inefficient in a large documents with complex structures and loose connections of relevant information that are hard to search and navigate. Consequently, they are especially penalized when the information being searched is hidden deeply. In this article, we introduce a partition-annotate-recommend (PAR) paradigm to improve the accessibility of digital libraries for visually impaired users. Our evaluation, involving the participation of visually impaired users, show that the RAIN system, built based on the PAR paradigm, reduces the navigational overhead significantly and enables visually impaired users to access complex digital libraries effectively.