Automated Construction of Web Accessibility Models from Transaction Click-streams

Published Version

Preview

PDF (2435Kb)

Abstract

Screen readers, the dominant assistive technology used by visually impaired people to access the Web, function by speaking out the content of the screen serially. Using screen readers for conducting online transactions can cause considerable information overload, because transactions, such as shopping and paying bills, typically involve a number of steps spanning several web pages. One can combat this overload by using a transaction model for web accessibility that presents only fragments of web pages that are needed for doing transactions. We can realize such a model by coupling a process automaton, encoding states of a transaction, with concept classifiers that identify page fragments “relevant” to a particular state of the transaction. In this paper we present a fully automated process that synergistically combines several techniques for transforming unlabeled click-stream data generated by transactions into a transaction model. These techniques include web content analysis to partition a web page into segments consisting of semantically related content, contextual analysis of data surrounding clickable objects in a page, and machine learning methods, such as clustering of page segments based on contextual analysis, statistical classification, and automata learning. The use of unlabeled click streams in building transaction models has important benefits: (i) visually impaired users do not have to depend on sighted users for creating manually labeled training data to construct the models; (ii) it is possible to mine personalized models from unlabeled transaction click-streams associated with sites that visually impaired users visit regularly; (iii) since unlabeled data is relatively easy to obtain, it is feasible to scale up the construction of domain-specific transaction models (e.g., separate models for shopping, airline reservations, bill payments, etc.); (iv) adjusting the performance of deployed models over time with new training data is also doable. We provide preliminary experimental evidence of the practical effectiveness of both domain-specific, as well as personalized accessibility transaction models built using our approach. Finally, this approach is applicable for building transaction models for mobile devices with limited-size displays, as well as for creating wrappers for information extraction from web sites.