Sentiment Analysis of Public Transport Feedback Using Twitter and BERT

Ibrahim, Mohammed; Nazar, Mustafa; Hussai, Sara

doi:10.25673/123731

Proceedings of International Conference on Applied Innovation in IT · 2026/03/31 · Vol. 14 · Issue 1 · pp. 1287–1293

Sentiment Analysis of Public Transport Feedback Using Twitter and BERT

Mohammed Fadhil Ibrahim, Mustafa Nazar, Sara Salam Ali and Salah Yehia Hussai

📄 Download PDF DOI: 10.25673/123731

Abstract

People are now more likely to talk about their experiences with public transportation on social media, so public transportation systems are being judged not only on how well they work but also on how happy their passengers are. This research introduces a BERT-based sentiment analysis pipeline designed to categorize Twitter feedback regarding bus and metro services into positive, neutral, and negative classifications. Using the Twitter API, data were collected from six metropolitan areas. Then, baseline models (TF-IDF + Logistic Regression, BiLSTM, DistilBERT) and fine-tuned BERT were trained on the data. BERT beats the baselines by getting the highest accuracy (0.81) and Macro-F1 (0.79) when compared to other models. The confusion matrix analysis showed that most of the mistakes happened between neutral and negative classes. This shows how subtle rider complaints can be. Robustness testing over six months showed that BERT was stable against temporal drift. SHAP-based interpretability, on the other hand, showed important tokens like "delay," "crowded," and "clean." This study substantiates BERT's efficacy for transport sentiment analysis and underscores the necessity for secure and ethical management of user data.

Keywords

Sentiment Analysis BERT Twitter Public Transport Macro-F1 Explainable AI.

References

M. M. Rahimi, E. Naghizade, M. Stevenson, and S. Winter, “Service quality monitoring in confined spaces through mining Twitter data,” Journal of Spatial Information Science, vol. 2020, no. 21, pp. 229-261, 2020.
C. Collins, S. Hasan, and S. V. Ukkusuri, “A novel transit rider satisfaction metric: Rider sentiments measured from online social media data,” Journal of Public Transportation, vol. 16, no. 2, pp. 21-45, 2013.
M. Pota, M. Ventura, R. Catelli, and M. Esposito, “An effective BERT-based pipeline for Twitter sentiment analysis: A case study in Italian,” Sensors, vol. 21, no. 1, p. 133, 2020.
S. Das and H. A. Zubaidi, “City transit rider tweets: understanding sentiments and politeness,” Journal of Urban Technology, vol. 30, no. 1, pp. 111-126, 2023.
T. El-Diraby, A. Shalaby, and M. Hosseini, “Linking social, semantic and sentiment analyses to support modeling transit customers’ satisfaction: Towards formal study of opinion dynamics,” Sustainable Cities and Society, vol. 49, p. 101578, 2019.
S. Kumar and R. Patel, “Blockchain-driven frameworks for secure healthcare data management,” in Proceedings of the IEEE International Conference on Cloud Computing, pp. 1-8, 2025, [Online]. Available: https://doi.org/10.1109/11015778.
J. Wang, L. Zhao, and Y. Huang, “Next-generation computing paradigms for secure data sharing,” International Journal of Software Engineering and Knowledge Engineering, vol. 35, no. 2, pp. 225-240, 2025, [Online]. Available: https://doi.org/10.1142/S0219649225500406.
Y. Mao, Q. Liu, and Y. Zhang, “Sentiment analysis methods, applications, and challenges: A systematic literature review,” Journal of King Saud University-Computer and Information Sciences, vol. 36, no. 4, p. 102048, 2024.
M. Rodríguez-Ibánez, A. Casánez-Ventura, F. Castejón-Mateos, and P. M. Cuenca-Jiménez, “A review on sentiment analysis from social media platforms,” Expert Systems with Applications, vol. 223, p. 119862, 2023.
Å. Jevinger, C. Zhao, J. A. Persson, and P. Davidsson, “Artificial intelligence for improving public transport: a mapping study,” Public Transport, vol. 16, no. 1, pp. 99-158, 2024.
A. Kinra, S. Beheshti-Kashi, R. Buch, T. A. Sick Nielsen, and F. Pereira, “Examining the potential of textual big data for public policy decision-making on driverless cars: A case study from Denmark,” Transport Policy, vol. 98, 2020.
M. A. Sayed, M. A. Hossain, M. M. Rahman, G. M. N. Ali, M. A. Islam, K. C. Paul, and X. Qin, “Public sentiment analysis of roadway work zones using social media data and machine learning models,” Data Science and Management, 2025.
S. Singh and A. Pareek, “Sentiment analysis on public transportation using different tools and techniques: A literature review,” in International Conference on Emerging Technologies in Computer Engineering, pp. 99-110, Springer, Cham, 2022.
H. F. Khelil, M. F. Ibrahim, H. A. Hussein, and R. K. Naser, “Evaluation of different stemming techniques on Arabic customer reviews,” Journal of Techniques, vol. 6, no. 2, pp. 1-8, Feb. 2024, [Online]. Available: https://doi.org/10.51173/jt.v6i2.2313.
L. Chaudhary, N. Girdhar, D. Sharma, J. Andreu-Perez, A. Doucet, and M. Renz, “A review of deep learning models for Twitter sentiment analysis: Challenges and opportunities,” IEEE Transactions on Computational Social Systems, vol. 11, no. 3, pp. 3550-3579, 2023.
E. J. Hadi, M. F. Ibrahim, and A. I. Mohammed, “Towards news classification, a machine learning approach based on stemming and feature extraction,” Journal of Information & Knowledge Management, vol. 24, no. 3, p. 2550045, May 2025, [Online]. Available: https://doi.org/10.1142/S0219649225500455.
Y. Zhang, H. Li, and X. Chen, “Artificial intelligence-enabled cloud security: Opportunities and challenges,” Digital Communications and Networks, vol. 11, no. 2, pp. 55-66, 2025, [Online]. Available: https://doi.org/10.1016/j.dcan.2025.01.00.
F. A. Bida and H. A. Naser, “Diagnostic of osteoporosis using backpropagation neural networks,” Journal of Techniques, vol. 7, no. 2, pp. 10-20, 2025, [Online]. Available: https://doi.org/10.51173/jt.v7i2.2597.
H. Traboulsi and M. I. Salem, “The role of electronic governance in enhancing entrepreneurial performance,” Technical Journal of Management Sciences, vol. 2, no. 1, pp. 13-20, 2025, [Online]. Available: https://doi.org/10.51173/tjms.v2i1.23.