Improving Web Accessibility: Computing New Web Page Design with NSGA-II for People with Low Vision

As society becomes increasingly aware of the need to take disabilities into account, new information technologies and intensive use of computers can be a chance or create new barriers. In the specific case of people with low vision, efforts to improve e-accessibility are mainly focused on the provision of third-party tools. Assistive technologies like screen magnifiers adapt graphical user interfaces to increase the quality of the perceived information. However, when these technologies deal with the Web, they are not able to meet all specific needs of people with low vision. In this paper, we propose an approach to make Web pages more accessible for users with specific needs. User preferences can concern font size, font family, text color, word and letter spacing, link color and decoration or even more complex features regarding brightness, relative size or contrast. We also take into account and encode the designer's graphical choices as designer preferences. Solving preferences of the user and of the designer to obtain a new Web page design is an optimization problem that we deal with Non-dominated Sorting Genetic Algorithm II (NSGA-II), a polynomial Multi-Objective Genetic Algorithm. We conducted detailed tests and evaluated the running time and quality of results of our tool on real Web pages. The results show that our approach for adapting Web page designs to specific user needs with NSGA II is worthwhile on real Web pages.

[1]  DebK.,et al.  A fast and elitist multiobjective genetic algorithm , 2002 .

[2]  Vicki L. Hanson,et al.  Web accessibility: a broader view , 2004, WWW '04.

[3]  Yuanyuan Zhang,et al.  Search-based software engineering: Trends, techniques and applications , 2012, CSUR.

[4]  Marco Laumanns,et al.  Combining Convergence and Diversity in Evolutionary Multiobjective Optimization , 2002, Evolutionary Computation.

[5]  Lothar Thiele,et al.  Multiobjective evolutionary algorithms: a comparative case study and the strength Pareto approach , 1999, IEEE Trans. Evol. Comput..

[6]  Krzysztof Z. Gajos,et al.  Automatically generating user interfaces adapted to users' motor and vision capabilities , 2007, UIST.

[7]  David E. Goldberg,et al.  A niched Pareto genetic algorithm for multiobjective optimization , 1994, Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence.

[8]  Ronen I. Brafman,et al.  Preference-Based Configuration of Web Page Content , 2001, IJCAI.

[9]  Jaideep Srivastava,et al.  Automatic personalization based on Web usage mining , 2000, CACM.

[10]  Melody Y. Ivory,et al.  Using Automated Tools to Improve Web Site Usage by Users with Diverse Abilities , 2003 .

[11]  David Corne,et al.  The Pareto archived evolution strategy: a new baseline algorithm for Pareto multiobjective optimisation , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[12]  Souhila Kaci,et al.  Working with Preferences: Less Is More , 2011, Cognitive Technologies.

[13]  Fernando Sánchez-Figueroa,et al.  On Adaptability of Web Sites for Visually Handicapped People , 2002, AH.

[14]  Borja Gamecho,et al.  Automatically generating tailored accessible user interfaces for ubiquitous services , 2011, ASSETS '11.

[15]  Mohamed Slimane,et al.  Improving Web Accessibility for Dichromat Users through Contrast Preservation , 2012, ICCHP.

[16]  Marianne Huchard,et al.  Web Page Personalization to Improve e-Accessibility for Visually Impaired People , 2014 .

[17]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..