Developing a blueprint for mobile data visualization

It is predicted that by 2025, almost three quarters of the internet users in the world will be mobile-only users.

While mobile devices provide ready access to data, there are limitations to how the data can be optimally presented due to the small form factor and limited screen size.

For example, it is a lot easier to show 10,000 data points on a desktop compared to a smartphone, which typically has a screen size of 2.82 inches (71.5 mm) by 5.78 inches (146.7 mm).

When data is presented on a small screen, it is either truncated, distorted or too busy for it to be legible—which is why mobile devices cannot display all the required information at once.

Mobile devices also do not have access to precise input methods like the mouse to 'click', 'drag and drop' or 'brush' for users to interact with the data.

Another limitation of mobile devices is the difficulty users face in browsing data.

"It is without a doubt that good mobile data visualization is critical for increasing user comprehension, pattern detection, monitoring and analysis," Assistant Professor of Computer Science, Wang Yong informed the Office of Research.

"Given the widespread use of mobile devices to monitor health vitals, read financial charts and decipher all sorts of trends for work and life, the field of mobile data visualization will become more and more important in years to come," he added.

"There are three important aspects that are critical to optimizing layout on mobile devices. One, enough data must be displayed for users to make sense of it. Two, the ability to zoom in and out, to provide the level of data granularity to the users. And three, there must be a way for users to select a subset of the data to extract the information that is needed," he explained.

"As a researcher, I also want to understand the impact of the different modes of interaction—speech, touch and tilting on enhancing user readability of data," he continued.

The research

This research that will be helmed by Professor Wang and Assistant Professor Kotaro Hara from SMU School of Computing and Information Systems is believed to be one of the world's pioneering research projects on mobile data visualization.

The research is designed in two work packages.

Work Package 1

In Work Package 1, 30 data visualization experts from both academia and industry (e.g., Microsoft Research and Tableau) will be interviewed to collect their feedback on the challenges and issues they are facing in the real-world design, creation, and applications of mobile data visualizations.

The team will also gather visualization examples from the internet—to understand and ascertain how the different visualization techniques such as scatter plot, bar chart, pie chart, etc. can be optimally displayed on mobile devices without losing the meaning of the data.

Additionally, the researchers will conduct an in-depth examination on the use of artificial intelligence (AI) on mobile data visualization. Particularly, how can AI seamlessly respond and translate data visualization from desktops to mobile devices?

At the end of this work package, the team intends to publish a comprehensive guide that details the principles and frameworks for desktop to mobile data visualization conversion.

Work Package 2

Work Package 2 is centered on the exploration of how the three user interactional modes—i. speech using natural language processing; ii. touch using fingers; iii tilting—can improve the user interactions with mobile data visualizations.

Speech recognition, using natural language processing (NLP), is a convenient way for users to give instructions to change the layout of data or extract a subset of the data. For example, if you want to see your step count before and after the COVID-19 pandemic, all you have to do is 'speak' to the data visualizations through voice.

While it is natural and intuitive to use touch, one key limitation is the large selection area to accommodate the size of a finger for data manipulation.

Tilting, which is a unique interaction input for mobile devices, will also be examined. In this part of the research, the team is investigating the ways by which mobile users can navigate and control animated transitions of data.

When these work packages are completed, Professor Wang's team intends to further the research by examining the combination of the different modalities to enhance data readability and user experience of data visualizations on mobile devices.

Professor Wang is excited about what could prove to be pioneering research: "It is so exhilarating to be working in this field of cutting-edge research. No blueprint currently exists for desktop-to-mobile conversation for data visualization. We aim to be the pioneers in pushing the boundaries for this research."

Provided by Singapore Management University