Abstract
The way users interact on social media can indicate their well-being. When depressed, people’s feelings tend to be more evident, affecting how users interact and demonstrating their feelings on social media. This paper presents a new approach for the temporal assessment of emotional behavior and interaction among depressed users on social networks. We start by modeling user interactions using complex networks, grouping users through time using the Clauset-Newman-Moore greedy modularity maximization. We evaluate the built networks using metrics such as assortativity, density, clustering, diameter, and shortest path length, closeness, and coverage. Then, we propose EMUS, a method for establishing an emotional user score based on the extraction of emotional features in texts of posts and comments. To extract emotional features, we combine the use of the Empath framework and VADER lexicon. Finally, based on the standard deviation among users, we establish a metric for assessing mood levels. We evaluated users for 33 days, and the results show a sequence of mixed emotional behaviors with high correlations between the number of active users in the network communities, and the form and quality of interactions. The developed approach can be further applied to other database graphs, for different sequential pattern analysis and text-mining contexts.
Highlights
O NLINE social networks have become an increasingly natural environment for sharing opinions and feeling, gaining importance in people’s daily lives
MATERIALS AND METHODS we detail methods and materials used in each stage of the process: data collection and description of users; the modulation of the user’s interactions of the social network through complex networks and the detection of communities; the metrics used to assess and respond to specific aspects of the interaction behavior among depressive users; the approaches used to extract topics discussed in the communities and to recognize feelings and emotional characteristics
EMUS works as follows: For each 3-day shift in time, we model the complex network of user interactions and calculate the user’s emotional score, extract contextual characteristics from the conversations, and identify the most common emotions that each user demonstrates through their posts and comments
Summary
O NLINE social networks have become an increasingly natural environment for sharing opinions and feeling, gaining importance in people’s daily lives. Different social factors and contexts are supported by the use of social networks, such as presented in collaborative works [1], learning [2], relationships [3], and dating [4]. It is common to first meet someone on a social network and afterward meet in person [5]. People have been included social media in their routine, whether for personal or professional use, causing a feeling that life does not have the same meaning without online interactions. The researchers in Affective Computing, an area of science formed by researchers of Computing and Emotions, have explored social media as another lens to model and evaluate emotional behaviors. In Affective Computing, several state-of-the-art technologies and methodologies for various
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
