Trust-Aware Kalman Filtering For Secure Traffic Prediction From Social Media Streams

Ahmad Faisal Abidin; Nor Surayati Mohamad Usop; Mohd Fadzil Abdul Kadir; Mohamad Afendee Mohamed

doi:10.29284/53karc50

Authors

Ahmad Faisal Abidin Department of Computer Science, Faculty of Informatics and Computing, Universiti Sultan Zainal Abidin, 22200 Besut, Terengganu, Malaysia
Nor Surayati Mohamad Usop Department of Computer Science, Faculty of Informatics and Computing, Universiti Sultan Zainal Abidin, 22200 Besut, Terengganu, Malaysia
Mohd Fadzil Abdul Kadir Department of Computer Science, Faculty of Informatics and Computing, Universiti Sultan Zainal Abidin, 22200 Besut, Terengganu, Malaysia
Mohamad Afendee Mohamed Department of Computer Science, Faculty of Informatics and Computing, Universiti Sultan Zainal Abidin, 22200 Besut, Terengganu, Malaysia

DOI:

https://doi.org/10.29284/53karc50

Keywords:

Kalman Filtering, Traffic Prediction, Data Security, Social Media Analytics, Twitter Traffic Data.

Abstract

The growing use of social media platforms, particularly Twitter, for real-time reporting of road traffic conditions presents new opportunities for intelligent transportation systems. However, traffic data derived from Twitter is inherently noisy, unstructured, and susceptible to misinformation, which poses significant challenges to prediction accuracy and data security. This review paper provides an analytical examination of Kalman Filter (KF)–based models for enhancing the reliability, robustness, and integrity of traffic information extracted from Twitter. The study focuses on variations of the Kalman Filter, with particular emphasis on the Extended Kalman Filter (EKF), and evaluates their effectiveness in handling uncertain, incomplete, and dynamically evolving data patterns typical of social media streams. Furthermore, the review discusses the role of KF-based prediction in supporting anomaly detection mechanisms for identifying malicious, manipulated, or bot-generated traffic reports, thereby improving data authenticity and trustworthiness. Existing studies indicate that Kalman Filtering techniques enhance traffic flow prediction accuracy, reduce noise in user-generated content, and improve model stability in rapidly changing traffic environments. Despite these advantages, several challenges remain, including scalability, semantic interpretation of textual data, and integration with cybersecurity frameworks such as sentiment analysis, machine learning–based classification, and blockchain-enabled validation. The paper concludes that Kalman Filter models hold strong potential for securing and predicting Twitter-sourced traffic data; however, further interdisciplinary research is required to develop optimized hybrid systems that integrate natural language processing, advanced filtering techniques, and robust security frameworks. The increasing reliance on social media platforms such as Twitter for real-time traffic monitoring presents both opportunities and challenges for intelligent transportation systems (ITS). While Twitter enables rapid dissemination of traffic-related information, the data is inherently noisy, unstructured, and vulnerable to misinformation and malicious manipulation. This paper proposes a secure Kalman Filter–based framework for trust-aware traffic prediction using Twitter-derived data. The proposed approach integrates simulated traffic data from SUMO with real-time social media observations through a data fusion mechanism governed by the Kalman Filter. The Extended Kalman Filter (EKF) is employed to model nonlinear traffic dynamics, while trust-aware filtering is achieved through adaptive Kalman Gain adjustment. Security mechanisms including digital signatures, blockchain validation, and anomaly detection are incorporated to enhance data integrity. Experimental results demonstrate improved prediction accuracy and robustness against unreliable inputs.

References

[1] W. Yao and S. Qian, “From Twitter to traffic predictor : Next-day morning traffic prediction using social media data,” Transp. Res. Part C, vol. 124, no. January, p. 102938, 2021, doi: 10.1016/j.trc.2020.102938.

[2] M. Zia, J. Fürle, C. Ludwig, S. Lautenbach, S. Gumbrich, and A. Zipf, “SocialMedia2Traffic : Derivation of Traffic Information from Social Media Data,” 2022.

[3] T. Mecheva and N. Kakanakov, “Cybersecurity in Intelligent Transportation Systems,” pp. 1–12, 2020.

[4] P. K. Putra, R. Mahendra, and I. Budi, “Traffic and road conditions monitoring system using extracted information from Twitter,” J. Big Data, 2022, doi: 10.1186/s40537-022-00621-3.

[5] E. Alomari, “Iktishaf : a Big Data Road-Traffic Event Detection Tool Using Twitter and Spark Machine Learning,” pp. 603–618, 2023.

[6] A. Essien, “A deep-learning model for urban traffic flow prediction with traffic events mined from twitter,” pp. 1345–1368, 2021.

[7] W. Yao and S. Qian, “From Twitter to traffic predictor: Next-day morning traffic prediction using social media data,” Transp. Res. Part C Emerg. Technol., vol. 124, no. January, p. 102938, 2021, doi: 10.1016/j.trc.2020.102938.

[8] K. Hayawi, S. Saha, M. Mehedy, M. Sujith, S. Mathew, and M. Kaosar, “Social media bot detection with deep learning methods : a systematic review,” Neural Comput. Appl., vol. 35, no. 12, pp. 8903–8918, 2023, doi: 10.1007/s00521-023-08352-z.

[9] W. Melhem, A. Abdi, and F. Meziane, “Traffic Detection and Forecasting from Social Media Data Using a Deep Learning-Based Model , Linguistic Knowledge , Large Language Models , and Knowledge Graphs,” vol. 2, no. Ic3k, pp. 235–242, 2024, doi: 10.5220/0013066900003838.

[10] A. F. Abidin, M. Kolberg, and A. Hussain, “Integrating SUMO and kalman filter models towards a social network based approach of public transport arrival time prediction,” Int. J. Simul. Syst. Sci. Technol., vol. 16, no. 3, pp. 5.1-5.9, 2015, doi: 10.5013/IJSSST.a.16.03.05.

[11] N. Deshpande and H. J. Park, “International Journal of Transportation Physics-informed deep learning with Kalman filter mixture for traffic state prediction q,” Int. J. Transp. Sci. Technol., vol. 17, pp. 161–174, 2025, doi: 10.1016/j.ijtst.2024.04.002.

[12] E. Management, N. Deshpande, and H. J. Park, “ODU Digital Commons Physics-Informed Deep Learning With Kalman Filter Mixture for Traffic State Prediction Physics-informed deep learning with Kalman filter mixture for traffic state prediction q,” pp. 161–174, 2025.

[13] K. Al Momin, S. Barua, M. S. Jamil, and O. F. Hamim, “Short duration traffic flow prediction using kalman filtering,” AIP Conf. Proc., vol. 2713, 2023, doi: 10.1063/5.0129721.

[14] M. Rajagopal et al., “An efficient intelligent transportation system for traffic flow prediction using meta-temporal hyperbolic quantum graph neural networks,” Sci. Rep., vol. 15, no. 1, pp. 1–21, 2025, doi: 10.1038/s41598-025-10794-5.

[15] Y. Wang, N. Masoud, and A. Khojandi, “Real-Time Sensor Anomaly Detection and Recovery in Connected Automated Vehicle Sensors,” pp. 1–11, doi: 10.1109/TITS.2020.2970295.Yiyang.

[16] H. Zhang, G. Yang, H. Yu, and Z. Zheng, “Kalman Filter-Based CNN-BiLSTM-ATT Model for Traffic Flow Prediction,” no. September 2022, 2023, doi: 10.32604/cmc.2023.039274.

[17] A. F. Abidin and M. Kolberg, “Towards Improved Vehicle Arrival Time Prediction in Public Transportation: Integrating SUMO and Kalman Filter Models,” Proc. - UKSim-AMSS 17th Int. Conf. Comput. Model. Simulation, UKSim 2015, pp. 147–152, 2016, doi: 10.1109/UKSim.2015.63.

[18] D. O. I. Link, A. F. Abidin, M. Fadzil, and A. Kadir, “Forecasting Road Traffic Using Kalman Filter Models Received Feb 2025 Accepted Aug 2025 Published Sep 2025,” vol. 59, no. 9, pp. 16–39, 2025.

[19] H. Zhang, R. Seshadri, A. A. Prakash, C. Antoniou, F. C. Pereira, and M. Ben-Akiva, “Improving the accuracy and efficiency of online calibration for simulation-based Dynamic Traffic Assignment,” Transp. Res. Part C Emerg. Technol., vol. 128, 2021, doi: 10.1016/j.trc.2021.103195.

[20] M. A. Makridis and A. Kouvelas, “An adaptive framework for real-time freeway traffic estimation in the presence of CAVs,” Transp. Res. Part C, vol. 149, no. February, p. 104066, 2023, doi: 10.1016/j.trc.2023.104066.

[21] N. Delhi, “TWITTER BASED TRAFFIC ANALYSIS AND PREDICTION MODEL FOR PLANNED EVENTS Department of Information Technology,” vol. 20, no. 10, pp. 2095–2112, 2021.

[22] B. Wang, P. Qian, Y. Cheng, Y. Qian, and J. Zhang, “A short ‑ term traffic flow prediction network using filter ‑ genetic feature selection method,” 2025.

[23] A. Emami, M. Sarvi, and S. Asadi Bagloee, “Using Kalman filter algorithm for short-term traffic flow prediction in a connected vehicle environment,” J. Mod. Transp., vol. 27, no. 3, pp. 222–232, 2019, doi: 10.1007/s40534-019-0193-2.

[24] A. K. Shafik and H. A. Rakha, “Real-Time Turning Movement, Queue Length, and Traffic Density Estimation and Prediction Using Vehicle Trajectory and Stationary Sensor Data,” Sensors, vol. 25, no. 3, 2025, doi: 10.3390/s25030830.

[25] F. Chen, Z. Chen, and C. Lu, “Graph Convolutional Networks with Kalman Filtering for Traffic Graph Convolutional Networks with Kalman Filtering for Traffic Prediction,” no. November 2020, 2026, doi: 10.1145/3397536.3422257.

[26] C. Shih and F. Hwang, “Data-driven numerical simulation with extended Kalman filtering and long short-term memory networks for highway traffic flow prediction,” J. Mech., vol. 40, no. December 2023, pp. 31–43, 2024, doi: 10.1093/jom/ufad046.

[27] Y. Muratoglu and A. Alkaya, “Unscented Kalman Filter based State of Charge Estimation for the Equalization of Lithium-ion Batteries on Electrical Vehicles,” Eng. Technol. Appl. Sci. Res., vol. 9, no. 6, pp. 4876–4882, 2019, doi: 10.48084/etasr.3111.

[28] D. López and A. Lozano, “Comparing the accuracy of real-time transit arrival estimations with a Kalman Filter motion model in different transit networks models,” Transp. Res. Procedia, vol. 78, no. 2023, pp. 111–118, 2024, doi: 10.1016/j.trpro.2024.02.015.