SignKG: Spatially aware urban business knowledge graph construction from Vietnamese street signboards
,
,
,
and
*
Article Sidebar
Full Text: 
PDF
Main Article Content
Abstract
Street-level signboards contain valuable urban business information, yet conventional OCR-based systems typically return isolated text strings without identifying their semantic roles or relationships. This limitation is particularly challenging for Vietnamese signboards because of decorative fonts, complex diacritics, irregular layouts, and ambiguous text content. This paper proposes SignKG, a spatially aware framework that transforms Vietnamese street-signboard images into structured urban business knowledge graphs. SignKG integrates visual perception, spatially aware entity extraction, constrained relation inference, and graph optimisation. The framework jointly exploits lexical semantics, textbox positions, neighbourhood context, business rules, and symbolic compatibility constraints to identify business entities and construct semantically consistent relations. Experiments on a real-world Vietnamese signboard dataset show that SignKG achieves entity-, relation-, and graph-level F1 scores of 0.910, 0.874, and 0.851, respectively. It also obtains a valid-edge ratio of 0.967 and a store-connectivity ratio of 0.951, outperforming rule-based, text-only, layout-aware, and LLM-based baselines. Ablation results confirm that spatial information is critical for entity disambiguation, while symbolic constraints improve graph validity and structural consistency.
Article Details
Conflict of Interest
© 2026 The authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License.
References
Belzarena, D., Mowlavi, S., Artola, A., Mariño, C., Gardella, M., Ramírez, I., Tadros, A., He, R., Bottaioli, N., Rajaei, B., et al. (2025). Improving OCR using internal document redundancy. In International Conference on Document Analysis and Recognition (pp. 244–260). Springer. https://doi.org/10.1007/978-3-032-04627-7_14
Chadzynski, A., Li, S., Grišiūtė, A., Chua, J., Hofmeister, M., Yan, J., Tai, H. Y., Lloyd, E., Tsai, Y. K., Agarwal, M., et al. (2023). Semantic 3D city interfaces-Intelligent interactions on dynamic geospatial knowledge graphs. Data-Centric Engineering, 4, Article e20. https://doi.org/10.1017/dce.2023.14
Cheewaprakobkit, P., Lin, C. Y., Lin, K. H., & Shih, T. K. (2023). Robust signboard detection and recognition in real environments. IEEE Transactions on Consumer Electronics, 69(3), 421–430. https://doi.org/10.1109/TCE.2023.3257288
Chen, Z., Zhang, Y., Fang, Y., Geng, Y., Guo, L., Chen, X., Li, Q., Zhang, W., Chen, J., Zhu, Y., et al. (2024). Knowledge graphs meet multi-modal learning: A comprehensive survey. arXiv. https://arxiv.org/abs/2402.05391
Cocos, A., & Callison-Burch, C. (2017). The language of place: Semantic value from geospatial context. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers (pp. 99–104). Association for Computational Linguistics. https://doi.org/10.18653/v1/E17-2016
Dagdelen, J., Dunn, A., Lee, S., Walker, N., Rosen, A. S., Ceder, G., Persson, K. A., & Jain, A. (2024). Structured information extraction from scientific text with large language models. Nature Communications, 15(1), Article 1418. https://doi.org/10.1038/s41467-024-45563-x
Deng, S., Wang, C., Li, Z., Zhang, N., Dai, Z., Chen, H., Xiong, F., Yan, M., Chen, Q., Chen, M., et al. (2023). Construction and applications of billion-scale pre-trained multimodal business knowledge graph. In 2023 IEEE 39th International Conference on Data Engineering (ICDE) (pp. 2988–3002). IEEE. https://doi.org/10.1109/ICDE55515.2023.00229
Dey, L. (2024). Knowledge graph-driven data processing for business intelligence. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 14(3), Article e1529. https://doi.org/10.1002/widm.1529
Li, P., Pei, B., Zhang, Y., Zhu, Q., Liu, T., Liu, J., & Du, P. (2026). SignSpotter: Scene text-informed cross-modal semantic consistency verification for street view images and POI data. International Journal of Applied Earth Observation and Geoinformation, 152, Article 105414. https://doi.org/10.1016/j.jag.2026.105414
Liu, Y., Yamamura, T., Tanaka, T., & Ohnishi, N. (2000). Extraction and distortion rectification of signboards in a scene image for robot navigation. IEEJ Transactions on Electronics, Information and Systems, 120(7), 1026–1034. https://doi.org/10.1541/ieejeiss1987.120.7_1026
Mazumder, T., Nusrat, F., Mahi, A. B. S., Brishty, J. B., Rahman, R., & Helaly, T. (2025). Automated and efficient Bangla signboard detection, text extraction, and novel categorization method for underrepresented languages in smart cities. Results in Engineering, 26, Article 105156. https://doi.org/10.1016/j.rineng.2025.105156
Nguyen, T. T. H., Jatowt, A., Coustaty, M., & Doucet, A. (2021). Survey of post-OCR processing approaches. ACM Computing Surveys (CSUR), 54(6), 1–37. https://doi.org/10.1145/3453476
Thanh, B. C., Hung, N. T., Dung, N. T., Tu, N. A., Van Anh, N., Quynh, L. T., & Duc, L. X. (2025). HOCR: A hybrid OCR approach to Vietnamese scanned administrative text recognition. In 2025 17th International Conference on Knowledge and System Engineering (KSE) (pp. 1–6). IEEE. https://doi.org/10.1109/KSE68178.2025.11309655
Wang, Y., Qiao, M., Liu, H., & Ye, X. (2018). Qualitative spatial reasoning on topological relations by combining the semantic web and constraint satisfaction. Geo-spatial Information Science, 21(2), 80–92. https://doi.org/10.1080/10095020.2018.1430659