A Tool to Examine a User's Emotions and Moods

Zhao, J., Gou, L., Wang, F., & Zhou, M. (2014). PEARL: An interactive visual analytic tool for understanding personal emotion style derived from social media. In Proceedings of the 2014 IEEE Symposium on Visual Analytics Science and Technology (pp. 203–212). doi:10.1109/VAST.2014.7042496

In this article, the authors describe the development and design of PEARL, “a timeline-based visual analytic tool that allows users to interactively discover and examine a person’s emotional contributions” (p. 202) over time through individuals’ social media text. Such data is valuable both in providing increased self-understanding for individuals as well as information to personalize services for businesses. PEARL contains an analytic engine that performs emotion analysis on social media text, identifying individuals’ basic emotions and moods. PEARL’s visualization component then turns this data into a timeline-based visual emotional profile that distinguishes individuals’ emotional style in three categories: emotional outlook, emotional volatility, and emotional resilience. This visualization was designed with three main goals: (a) to summarize the personal emotions at multiple levels and perspectives, (b) to reveal emotional patterns over time, and (c) to provide data for analytical reasoning (p. 207). To test PEARL, the authors first compared a sample of users’ self-labeled tweets from a 1-month period to PEARL’s emotion analysis models. The authors then conducted an interview study to collect six users’ self-reflection as well as a Mechanical Turk survey with 50 workers to evaluate the system’s usefulness. The authors concluded that the system effectively supported users’ emotion analysis.  

‍