Taming data complexity in lifelogs

Abstract

As people continue to adopt technology-based self-tracking devices and applications, questions arise about how personal informatics tools can better support self-tracker goals. This paper extends prior work on analyzing and summarizing self-tracking data, with the goal of helping self-trackers identify more meaningful and actionable findings. We begin by surveying physical activity self-trackers to identify their goals and the factors they report influence their physical activity. We then define a cut as a subset of collected data with some shared feature, develop a set of cuts over location and physical activity data, and visualize those cuts using a variety of presentations. Finally, we conduct a month-long field deployment with participants tracking their location and physical activity data and then using our methods to examine their data. We report on participant reactions to our methods and future design opportunities suggested by our work.

Staying the Course

“…quit smoking, lose weight, etc.) [2,9]. It may also be influenced by individual characteristics such as intrinsic or extrinsic motivation, [31] or self-efficacy (e.g.…”

“…Personal and situational factors can also cause people to lapse from their ongoing goal pursuit. These include loss of motivation or reduced novelty [34] or external factors such as a change in work schedule, weather, travel, or injury [9].…”

“…Thus, relapse prevention methods should make it clear that such mistakes are common and not a sign of personal failure [22]. In other words, a slip "need not be a fall" [9]. As described earlier, one challenge of lapse management is that, on one hand, the negative impact of a lapse should be managed such that it does not lead to quitting.…”

“…The most common techniques in this category focus on designing tools that support goal setting, in particular the variety of goals that people might have [19,30]. Research also suggests that technology should support revision of goals, in particular as behavior-change goals change [9]. Research also proposed that technologies support the creation of action plans that are designed for the context, routines and preferences of the person using them [7,27].…”

Sigchi

“…Huron et al have proposed a new paradigm to create flexible, and tangible visualizations using physical building blocks that are dynamic and enable non-experts to create novel visualizations [3]. Another work by Epstein et al, titled "Taming Data Complexity in Lifelogs", defines cuts or subsets of collected data with shared features such as location or physical activity, and visualizes those cuts using a variety of presentation techniques including graphs, tables, and Sankey diagrams [4].…”