Many users have topics that they follow over time, whether for personal or professional reasons. The user has an ongoing interest in understanding what changes are happening in their overall topic area. This information retrieval task, which I will refer to as the “change detection” task, involves showing the user the latest developments within subtopics on their ongoing topic of interest. While systems exist to answer ad hoc questions, or provide general overviews of events that are happening in the world, few systems appear to focus on meeting the standing information needs of users—including experts—who follow a specific topic over time.
As part of my dissertation research, I ran a survey on users’ sort order preferences in social media to understand how they prefer to have feeds organized for change detection and other use cases. I am building upon this previous study to design and implement a prototype search system, known as DAYBREAK, based upon user feedback. For this next stage, I will perform an in-person user study focused on DAYBREAK, to determine the extent to which the user is able to complete change detection tasks successfully.