Automatic identification and role detection of visual elements in web pages

Akpınar, Mehmet Elgin
Web pages typically include many visual elements to support interaction with the user. However, if visual elements do not comply with web accessibility guidelines, they become inaccessible in alternative presentations, such as audio. This study presents an automatic role detection approach to identify visual elements in web pages and their roles. The system architecture has three major components: automatic identification of visual elements in web pages; automatic generation of heuristic rules from the knowledge base; and application of these rules for automatic annotation of visual elements with their roles. Moreover, automatic role detection approach is deployed in a case study of web page transcoding based on eye tracking. This thesis first explains the system architecture in detail and then presents both technical and user evaluations of the proposed approaches. User evaluation shows that the automatic role detection approach has around 80% receptive accuracy, but the proposed knowledge base could be further improved for better results. User evaluation also shows that the transcoding application partially improves the information quality of web pages.
Citation Formats
M. E. Akpınar, “Automatic identification and role detection of visual elements in web pages,” M.S. - Master of Science, Middle East Technical University, 2014.