Data visualization and interaction on smartwatch small screens

Thumbnail Image
Neshati, Ali
Journal Title
Journal ISSN
Volume Title
In this thesis, I will investigate the optimization of a small display while presenting graphic data such as line charts. Due to the small screen of smartwatches presenting the data could be challenging. To overcome these challenges, I will propose two techniques, Compression and Simplification, to improve the visualization techniques on smartwatches. The last part of this thesis is about interaction with visualization techniques on smartwatches. I will show how interacting with the smartwatch bezel and using some specific parts of the smartwatch display can be helpful to interact with various types of charts and graphs on smartwatches. In the first part of my thesis, I propose G-Sparks, a compact visual representation of glanceable line graphs for smartwatches. My exploration primarily considered the suitable compression axes for time-series charts. In a first study, I examine the optimal line-graph compression approach without compromising perceptual metrics, such as slope or height detection. I evaluated compressions of line segments, the elementary unit of a line graph, along the x-axis, y-axis, and xy-axes. Contrary to intuition, I find that condensing graphs yield more accurate reading of height estimations than non-compressed graphs, but only when these are compressed along the x-axis. Building from this result, I study the effect of an x-axis compression on users' ability to perform "glanceable" analytic tasks with actual data. Glanceable tasks include quick perceptual judgements of graph properties. Using biometric data (heart rate), I find that shrinking a line graph to the point of representing one data sample per pixel does not compromise legibility. As expected, such type of compression also has the effect of minimizing the needed amount of flicking to interact with such graphs. From the results, I offer guidelines to application designers needing to integrate line charts into smartwatch apps. Multiple embedded sensors enable smartwatch apps to amass large amounts of interrelated time-series data simultaneously, such as heart rate, oxygen levels or steps walked. Visualizing multiple interlinked datasets is possible on smartphones but remains challenging on small smartwatch displays. I propose a new technique, the Space-Filling Line Graph (SF-LG), that preserves the key visual properties of time-series graphs while making available space on the display to augment such graphs with additional information. Results from the first study (N=30) suggest that, while SF-LG makes available additional space on the small display, it also enables effective (i.e. quick and accurate) comprehension of key line graph tasks. I next implement a greedy algorithm to embed auxiliary information in the most suitable regions on the display. In a second study (N=27), I find that participants are efficient at locating and linking interrelated content using SF-LG in comparison to two baselines approaches. I conclude with guidelines for smartwatch space maximization for visual displays. I present BezelGlide, a novel suite of bezel interaction techniques designed to minimize screen occlusion and `fat finger' effects when interacting with common graphs on smartwatches. To explore the design of BezelGlide, I conducted two user studies. First, I quantified the amount of screen occlusion experienced when interacting with the smartwatch bezel. Next, I designed two techniques that involve gliding the finger along the smartwatch bezel for graph interaction. Full BezelGlide (FBG) and Partial BezelGlide (PBG) use the full or a portion of the bezel, respectively, to reduce screen occlusion while scanning a line chart for data. In the common value detection task, I find that PBG outperforms FBG and Shift, a touchscreen occlusion-free technique, both quantitatively and subjectively, also while mobile. I finally illustrate the generalizability potential of PBG to interact with common graph types making it a valuable interaction technique for smartwatch users. As the last piece of this thesis, I will show how some specific segments of the smartwatch display can be used as the interaction area to interact with multiple graphs on smartwatches. In the first experiment, I will investigate the occlusion area while the smartwatch user interacts with the whole smartwatch display. The result of this study will help to pick the right segments with minimum occlusion to interact with multiple graphs. For the interaction technique, the challenge will be mapping the selected segments (interaction area) with the area representing content.
Data visualization, Space-efficient visualizations, Interaction techniques, Smartwatches
Neshati A, Sakamoto Y, Leboe-Mcgowan L, Leboe-McGowan J, Serrano M, Irani P. G-sparks: Glanceable sparklines on smartwatches. In45th Conference on Graphics Interface (GI 2019) 2019 May 28 (pp. 1-9).
Neshati A, Sakamoto Y, Irani P. Challenges in Displaying Health Data on Small Smartwatch Screens. InITCH 2019 Jan 1 (pp. 325-332)
Neshati A, Rey B, Ahmed Sharif F, Bardot S, Latulipe C, Irani P. BezelGlide: Interacting with Graphs on Smartwatches with Minimal Screen Occlusion. InProceedings of the 2021 CHI Conference on Human Factors in Computing Systems 2021 May 17.
Neshati A, Alallah F, Rey B, Sakamoto Y, Serano M, Irani P. SF-LG: Space- Filling Line Graphs for Visualizing Interrelated Time-series Data on Smart- watches. submitted to In23rd International Conference on Human-Computer Interaction with Mobile Devices and Services 2021 Oct 5