Semi-Automated Tracking: A Balanced Approach for Self-Monitoring Applications

Abstract

We present an approach for designing self-monitoring technology called semi-automated tracking, which combines both manual and automated data collection methods. Through this approach, we aim to lower the capture burdens, collect data that is typically hard to track automatically, and promote awareness to help people achieve the goals of self-monitoring. We first specify three design considerations for semi-automated tracking-(1) data capture feasibility; (2) purpose of self-monitoring; and (3) motivation level. We then provide examples of semi-automated tracking applications in the domains of sleep, mood, and food tracking to demonstrate strategies we have developed to find the right balance between manual tracking and automated tracking, combining each of their benefits while minimizing their associated limitations.

Flexible and Mindful Self-Tracking

“…Notably, some people stop using self-tracking technologies after having gained new insights or developed new routines [17], and a few people lapse and resume self-tracking because of shifting life priorities [22,38,53]. Considering these circumstances and limitations of current personal informatics tools, recent research has highlighted the importance of supporting personalisation [29], customisation [30], and flexibility [10,36].…”

“…Recently, there has been a push to semiautomated [10] and flexible self-tracking [36] to overcome the limitations of current self-tracking technologies. However, there is a lack of dedicated research on how individuals use and especially design paper notebooks to engage in self-tracking [20], and what this understanding might imply for the design of flexible self-tracking systems.…”

“…For example, prior works suggest that people abandon consumer health technologies over time because of a lack of personally meaningful insights [22,38], and switch to paper notebooks to avoid unintended effects and to overcome technological boundaries [3,33]. Addressing these issues, recent research has proposed leveraging flexible self-tracking to account for people's individual and changing self-tracking practices [10,30,36].…”

Exploring Symmetric and Asymmetric Bimanual Eating Detection with Inertial Sensors on the Wrist

“…Automatically detecting eating activities is a cornerstone of a wide range of health applications, helping behavioral researchers understand the link between diet and disease [Hatori et al 2012], and enabling new forms of dietary self-monitoring such as semi-automated food journaling [Choe et al 2017]. Over the last few years, computing researchers have developed new approaches for automatic eating detection by making use of the inertial sensing capabilities in off-the-shelf devices such as mobile phones, smart watches, activity trackers, and wearable devices [Dong et al 2013, Thomaz et al 2015, Junker et al 2008, Amft and Tröster 2009, Merck et al 2016, Rahman et al 2015, Rahman et al 2016].…”

Toward Increasing Engagement in Substance Use Data Collection: Development of the Substance Abuse Research Assistant App and Protocol for a Microrandomized Trial Using Adolescents and Emerging Adults (Preprint)

“…However, detecting substance use with sensors is new and requires further validation; Bae et al [34] used mobile phone data to detect drinking episodes; SCRAM [35,36] and BACTrack Skyn [37] are wearable sensors that can continuously measure blood alcohol levels. Such sensors do not exist for other substances, however, and important correlates of substance use such as subjective stress level or mood cannot yet be reliably detected with sensors [38]. Therefore, manual data capture remains a valuable way to obtain substance use-related data.…”