Home // SOTICS 2021, The Eleventh International Conference on Social Media Technologies, Communication, and Informatics // View article

Community Interaction Optimization on Twitter for people with Mood Disorders

Authors:
Yuichi Okada
Naoya Ito
Tomoko Yonezawa

Keywords: Twitter, SNS, data mining, combinatorial optimization

Abstract:
This paper proposes our designed system to estimate the optimal social networking connections for people with mood disorders, such as depression. We collected data from Twitter and analyzed users' characteristics by adopting an emotional polarity value index. Based on these data analyses, we defined each user's positivity level and estimated the level of mood disorder from the content of their tweets. We also simulated the system's use by people with severe mood disorders. A computational model based on a knapsack problem, a combinatorial problem to solve using an optimization method, was created based on the hypothesis that people likely to have mood disorders will more likely connect with people who have similarly severe mood disorders. The proposed system solved it using an approximate solution method that can be computed in a practical amount of time. As a result, (1) the user preferred a user with the same mood disorder severity when a user connected to a user with frequent tweets and (2) the difference in mood disorder severity was not as important when the user connected to a user with infrequent tweets.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2021

Publication date: October 3, 2021

Published in: conference

ISSN: 2326-9294

ISBN: 978-1-61208-899-0

Location: Barcelona, Spain

Dates: from October 3, 2021 to October 7, 2021