Ph.D. Thesis Defense: Conor Kelton 'Towards a More User-Centered Web Experience'

Dates: 
Friday, March 5, 2021 - 2:00pm to 3:00pm
Location: 
Zoom - contact events@cs.stonybrook.edu for Zoom info.
Event Description: 
Talk abstract:

Studies from major players in the Web ecosystem, such as Amazon, Walmart, and Google, show that small improvements to Web page load performance can have an enormous downstream impact on revenue and end-user satisfaction. However, we posit common metrics used to quantify Web loads fail to account for the user perception. As users are unlikely to interact with the page until they perceive it to be loaded, we design a metric to capture when such a time occurs, called the user's perceived page load time. We use crowdsourcing to collect the users’ perceived page load time across ~600 desktop and mobile users as they view over 120 Web pages. For desktop browses, we asked users to respond with a simple mouse click when they thought the page load was complete. On mobile browsers,  the important content is spread across multiple viewports, causing users to scroll during the page load. So, we create a specialized user study to collect ground truth user perceived load time for mobiles while allowing users to scroll.

 

Our key finding is that across desktop and mobile browsers, metrics used to measure page load times do not correlate with the user-perceived page load times. As improving the user perception is critical, we next explore how to modify pages to actually reduce the user's perceived page load time. To do so, we leverage gaze tracking, which is commonly employed in the cognitive science literature to measure a user's focused attention. Our gaze study shows that certain objects on the page are more important than others  based on the collective fixation across users. We develop WebGaze, a system which prioritizes bandwidth to objects which demand user attention over those that do not during page loads. We show WebGaze actually improves the user perceived page load over state-of-the-art optimizations with real crowdsourced studies. In addition, we increase the reach of systems, like WebGaze, which rely on or improve the perceived load by developing models that can predict it as a function of readily available page load metrics.

 

Finally, we measure the user’s perception of Web pages implicitly by understanding how much they engage with pages. We measure engagement implicitly using their eye movements or mouse tracks without requiring explicit feedback or in-lab studies. As users of the live Web view page contents as they actually load, we use the user’s engagement to show how differing load orders of objects can result in very different levels of engagement for users, both during the page load process and after the page has already loaded. Our results suggest that developers can reorient the load orders of their pages to maximize how much users engage with page contents, thereby offering another potential methodology to improve the user's quality of experience with Web pages.

Computed Event Type: 
Mis