Yargı, Özgün and Varol, Onur (2026) #Boykot: brand-based political polarization [#Boykot: markalar üzerinden siyasi kutuplaşma]. In: 34th Signal Processing and Communications Applications Conference (SIU), Istanbul, Turkiye
Full text not available from this repository. (Request a copy)
Official URL: https://dx.doi.org/10.1109/SIU71813.2026.11637021
Abstract
This study examines Instagram discussions around #boykot as a case of digital activism and political polarization in Turkey following March 19, 2025. Using the hashtags #boykot, #boykotadevam, and #boykotadevamturkiyem, we collected 51,610 posts from 14,588 users between March 27, 2025 and September 24, 2025. We combine temporal analysis, sentiment classification, and hashtag co-occurrence network analysis. Sentiment labels are generated with TurkishBERTweet, a model adapted to Turkish social media language. Results show that domestically focused boycott narratives are strongly event-driven, peaking rapidly and then declining over time, while negative discourse toward brands associated with the Israel-Palestine conflict is more stable and persistent. Network findings further reveal a structurally polarized communication space with two salient poles, alongside a central area dominated by engagement-oriented and agenda-agnostic content. Overall, #boykot functions not only as a consumer boycott signal but also as a multi-layered arena where political identities, grievances, and polarization dynamics become visible.
| Item Type: | Papers in Conference Proceedings |
|---|---|
| Uncontrolled Keywords: | boycott; digital activism; network analysis; political polarization; sentiment analysis; Social media analytics |
| Divisions: | Faculty of Engineering and Natural Sciences |
| Depositing User: | Onur Varol |
| Date Deposited: | 09 Sep 2026 14:57 |
| Last Modified: | 09 Sep 2026 14:57 |
| URI: | https://research.sabanciuniv.edu/id/eprint/54468 |

