Personalized accessibility based Re-ranking of Search Engine Results (PARSER)

The relevancy and usability of the contents retrieved are the two major factors for deciding the utility value of web resources. In the case of persons with disabilities the usability aspect of the interface is solely dependent on the accessibility of the web contents. This paper proposes a model termed PARSER - Personalized Accessibility based Re-ranking of Search Engine Results which aims towards quantizing the optimality of the pages with respect to the user's personalized accessibility and information requirement context. The PARSER model proposes a metric termed PAS - Personalized Accessibility Score which computes the accessibility score of a given web page for a particular user profile. The model incorporates aggregation of three ranked lists derived by search engines, accessibility scores and personalized information requirement context, to build the re-ranked list. The interface proposed by the PARSER model is targeted towards reducing the information access barrier.