What do users complain about? A BERTopic study on Turkish e-commerce app reviews
ELECTRONIC COMMERCE RESEARCH, 2026 (SSCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1007/s10660-026-10182-9
- Dergi Adı: ELECTRONIC COMMERCE RESEARCH
- Derginin Tarandığı İndeksler: Social Sciences Citation Index (SSCI), Scopus, IBZ Online, ABI/INFORM, Aerospace Database, Compendex, INSPEC, zbMATH, Business Source Ultimate (EBSCO), Pharma Collection (ProQuest), Technology Collection (ProQuest)
- Karadeniz Teknik Üniversitesi Adresli: Evet
Özet
This study investigates the primary sources of user dissatisfaction in T & uuml;rkiye's leading e-commerce applications by analyzing user-generated content through advanced topic modeling techniques. As mobile commerce becomes increasingly integrated into daily life, user reviews offer valuable insights into both functional performance and perceived service quality. Yet, the vast and unstructured nature of these reviews poses challenges for conventional analysis methods. To address this, the BERTopic algorithm which combines contextual embeddings with class-based TF-IDF was applied to 18,685 Turkish-language user reviews collected from the Google Play Store using the google-play-scraper Python library. The analysis revealed recurring concerns around delivery delays, ineffective customer service, and technical issues related to payment and order tracking. The recurrence of these themes across different platforms reflects common patterns in key areas that influence user experience within T & uuml;rkiye's digital commerce ecosystem. By offering a structured and semantically coherent thematic mapping of user sentiment, this study provides actionable insights for improving service quality and contributes methodologically to natural language processing research on morphologically rich languages like Turkish.