Supporting Mood Introspection from Digital Footprints

Doing and Feeling: Relationships Between Moods, Productivity and Task-Switching

“…This detection system has an accuracy of 70% and kappa score around 0.45. Further, our previous studies [12], [13] showed that digital footprints are also useful to predict mood. In this study, we attempted to explore if the mood prediction accuracy will be improved by adding lifestyle or physical context information.…”

“…Past studies have used them to detect stress [32] and personality traits [33]. We aim to build a mood detection system using the digital duration features mentioned in [12] and include lifestyle factors as new features. We hypothesise that digital behaviour and lifestyle information could be useful to predict mood.…”

“…From the 1,244 mood self-reports data that we have collected, we investigated the users' digital activities 30 and 60 minutes prior to the mood self-report log, as recorded by RescueTime. We then extracted features from these RescueTime data based on our previous study that focused on duration features [12]. The extracted features consist of digital category duration percentage, distinct category number and productivity.…”

“…These two algorithms are selected since they are ensemble learning methods, which combine different individual models and tend to have less bias and variance. Similar to our previous study [12], the last 30% self-reports of each user were used for testing datasets to evaluate the models' performances. The evaluation metrics are accuracy, true positive rate (TPR), true negative rate (TNR), precision, and F-measure (F1) score.…”

“…Though these productivity tools are available to the public (e.g., RescueTime and ManicTime), based on our knowledge, only a few studies have utilised them [11]. In our previous studies, we have attempted to use their data to detect mood [12], [13]. We argue that these digital tool usage records contain rich information useful as machine learning features.…”