Assessing the Impact of Spatial Data Quality on Deep Learning Models for Travel Time Estimation

• Diêgo de A. Correia Universidade Federal de Alagoas (UFAL)
• Rafael Luciano L. Silva Universidade Federal de Alagoas (UFAL) https://orcid.org/0009-0000-0946-0909
• Vinicius Santiago Universidade Federal de Alagoas (UFAL)
• Jose Matheus S. Alves Universidade Federal de Alagoas (UFAL)
• Ruan T. Melo Universidade Federal de Alagoas (UFAL)
• J. Rui R. S. Fernandes Universidade Federal de Alagoas (UFAL)
• Fábio J. Coutinho Universidade Federal de Alagoas (UFAL)

DOI: https://doi.org/10.5753/sbbd.2026.249667

Resumo

Data quality is fundamental to the reliability of predictive models. In the context of spatial data, understanding how quality dimensions influence robustness is critical for intelligent transportation systems. This study investigates the impact of spatial data quality on Deep Learning models for Travel Time Estimation (TTE). The methodology encompasses the steps of data pollution, model training, and testing under two complementary scenarios: polluted training and test data. We evaluate three models using the Porto Taxi dataset while considering the impact of the Positional Accuracy and Completeness quality dimensions defined in ISO 19157-1:2023. The results indicate that the impact of data quality degradation varies according to both model architecture and application phase.

Palavras-chave: Data Quality for AI, Travel Time Estimation, Data Pollution, Spatial Data Quality, Data Quality, Deep Learning

Referências

Chen (2025). Exploration of intelligent geographic information system development in big data era. Journal of Computer, Signal, and System Research, 2(3):24–30.

Deng et al. (2022). Efficient trajectory similarity computation with contrastive learning. In Proceedings of the 31st ACM International Conference on Information & Knowledge Management (CIKM ’22), page 365–374. Association for Computing Machinery.

Hiniduma et al. (2025). Data readiness for AI: A 360-Degree Survey. ACM Computing Surveys, 57(9).

Jiang et al. (2023). Self-supervised trajectory representation learning with temporal regularities and travel semantics. In 2023 IEEE 39th International Conference on Data Engineering (ICDE), pages 843–855.

Li et al. (2018). Deep representation learning for trajectory similarity computation. ICDE 2018, pages 617–628.

Minghini et al. (2019). Openstreetmap history for intrinsic quality assessment: Is OSM up-to-date? Open Geospatial Data, Software and Standards, 4(1):1–17.

Mohammed et al. (2025). The effects of data quality on machine learning performance on tabular data. Information Systems, 132:102549.

Si et al. (2023). Data quality analysis and improvement: A case study of a bus transportation system. Applied Sciences, 13(19).

Silva et al. (2025). QualiBus: implementando métricas de qualidade para dados de transporte coletivo. SBBD 2025, pages 589–601. SBC.

Tavra et al. (2026). Determining the data quality-influencing characteristics of participants in volunteered geographic information campaigns – the case of maritime disasters. Geomatics, Natural Hazards and Risk, 17(1):2659907.

Vitagliano et al. (2023). Pollock: A data loading benchmark. VLDB, 16(8):1870–1882.

Wang et al. (2018a). When will you arrive? estimating travel time based on deep neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32.

Wang et al. (2018b). Learning to estimate the travel time. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD ’18), page 858–866. Association for Computing Machinery.

Zheng et al. (2023). Data-driven methods for travel time estimation: A survey. ITSC 2023, pages 1292–1299. ISSN: 2153-0017.

Zhou et al. (2024). Red: Effective trajectory representation learning with comprehensive information. Proc. VLDB Endow., 18(2):80–92.

Zuo et al. (2026). TriDGNet: Triple feature encoder-based dual granularity graph learning network for enhanced travel time estimation. IEEE Transactions on Intelligent Transportation Systems, 27(1):1606–1620.