Exploring the Use of Online Social Network Activity and Smartphone Photography as an Intervention to Track and Influence Emotional Well-Being

The proliferation of internet and mobile technologies has expanded the means of detecting and influencing mental health, with this thesis focusing on the affective phenomena associated with emotional well-being including mood, affect and emotion. Traditional detection techniques including surveys and self-reports are grounded in the psychological literature; however, they introduce an inhibiting burden on the participants. The ability to passively detect psychological state using technologies including online behavioural tracking and mobile sensors is a prevalent focus of the current literature. Traditional positive psychology interventions commonly involve emotionally expressive writing tasks which can also be tedious for participants. Augmenting traditional intervention techniques with technologies such as smartphone applications can be one method to modernise interventions. The first research study in this thesis aimed to utilise online social network (OSN) activity to detect mood changes. The study involved collecting the participants' behavioural activities such as likes, comments and tweets from their Facebook and Twitter profiles. Machine learning was used to create an algorithm to classify participants according to their online activity and their self-reported mood as ground truth. The findings indicated that participants can be grouped into those who displayed positive, negative or weak correlations with their online activity. Following the classification, the system used a sliding window of 7 days to track the participant's mood changes for those in the positive and negative groups. The second research study introduced a positive psychology intervention in the form of a smartphone application called SnapAppy which promotes positive thinking by integrating momentary smartphone photography with traditional intervention methodologies. Participants were required to take photos and write about positive moments, past events, acts of kindness and gratuitous situations, encouraging them to think more positively. The results indicated that features such as the number of photos taken, the effort applied to annotating the photos, the number of photos revisited and the photos containing people were positively correlated with an improvement in mood and affect. The product of this thesis is a novel method of passively tracking mood changes using online social network activity and an innovative smartphone intervention utilising photography to influence emotional well-being