Abstract
Content on the internet is often not accessible to all users as it has to be speciallyenhanced in order to be usable with assistive technologies. Particularly in the case of
information visualizations, blind and severely visually impaired people face the problem
that the displayed data can either not be explored at all or only with great difficulty, using screen readers and the keyboard. This often means that these users do not even
try to use the visualizations.
The aim of this master’s thesis is to develop a concept that enables screen reader users
to explore embedded information visualizations on the internet in a targeted manner.
The concept should enable users to get a complete overview of the data and its development as well as to search for specific data points. In addition, sonification is to be
integrated to support users in exploring the data.
To create the concept, the current state of science and technology is beeing researched
in order to incorporate existing findings. In addition, a preliminary study is conducted
with sighted users in order to identify the attributes of information visualization that
are meaningful and helpful for users. Discussions with screen reader users should reveal
the current problems and needs of the target group.
The developed concept will then implemented using Angular, TypeScript and D3.js for
bar charts, line charts and scatter plots. A user study will clarify whether the concept
and the prototype offer the screen reader users the hoped-for possibilities. The participants in the study will perform various tasks of visualization will be tested with different data sets.
The results of the user study were evaluated with regard to the feasibility of the tasks.
The interactions used by the participants to solve the tasks were also analyzed. The
study showed that the concept and the prototypes enable users to explore the data independently and in a targeted manner and to obtain an overview of the data progression
through sonification. However, the concept does not yet work as effectively for scatter plots as it does for bar and line charts.
| Date of Award | 2024 |
|---|---|
| Original language | German (Austria) |
| Supervisor | Mandy Keck (Supervisor) |