Cultural Differences in Signed Ego Networks on Twitter: An Investigatory Analysis
Abstract
Human social behaviour has been observed to adhere to certain structures. One such structure, the Ego Network Model (ENM), has been found almost ubiquitously in human society. Recently, this model has been extended to include signed connections. While the unsigned ENM has been rigorously observed for decades, the signed version is still somewhat novel and lacks the same breadth of observation. Therefore, the main aim of this paper is to examine this signed structure across various categories of individuals from a swathe of culturally distinct regions. Minor differences in the distribution of signs across the SENM can be observed between cultures. However, these can be overwhelmed when the network is centred around a specific topic. Indeed, users who are engaged with specific themes display higher levels of negativity in their networks. This effect is further supported by a significant negative correlation between the number of "general" topics discussed in a network and that network's percentage of negative connections. These findings suggest that the negativity of communications and relationships on Twitter are very dependent on the topics being discussed and, furthermore, these relationships are more likely to be negative when they are based around a specific topic.
- Publication:
-
arXiv e-prints
- Pub Date:
- May 2023
- DOI:
- 10.48550/arXiv.2305.10396
- arXiv:
- arXiv:2305.10396
- Bibcode:
- 2023arXiv230510396T
- Keywords:
-
- Computer Science - Social and Information Networks
- E-Print:
- This work was partially supported by the following projects: H2020 HumaneAI-Net (952026), SoBigData++ (871042), CHIST-ERA-19-XAI010 SAI, Partenariato Esteso PE00000013 - "FAIR", Centro Nazionale CN00000013 - "ICSC". In Companion Proceedings of the ACM Web Conference 2023