Urban Perception Extraction from Texts Shared on Social Media: Framework and Applications

• Frances Albert Santos Unicamp http://orcid.org/0000-0002-0110-6507
• Thiago Henrique Silva UTFPR
• Leandro Aparecido Villas Unicamp

DOI: https://doi.org/10.5753/sbrc_estendido.2025.6880

Abstract

This thesis presents an automatic, generic framework for extracting urban perceptions from Location-Based Social Network (LBSN) data. The framework is organized into five key layers: Data Collection, Preprocessing and Embeddings, Model Training, Knowledge Extraction, and Applications. By leveraging deep learning techniques, including advanced sentence embedding methods, the framework captures both lexical and semantic nuances in textual data, thereby efficiently extracting user perceptions of urban environments. This approach eliminates the need for labor-intensive field surveys and manual data extraction, allowing scalable real-time analysis. We validated the framework by applying it to selected urban areas in Chicago, New York City, and London, demonstrating its effectiveness in uncovering valuable insights about urban perceptions. Furthermore, a comparative evaluation using a public dataset derived from volunteers’ perceptions in a controlled experiment revealed a high level of agreement between the two sets of results. As a proof-of-concept, we introduce Real-Estate Urban Perceptions (REAL-UP), an innovative tool designed to enhance the real estate marketplace. REAL-UP provides interactive 2D maps that integrate traditional real-estate data (e.g., rent prices and property types) with enriched information on neighborhood emotions, sentiments, and brief narrative reviews generated by a Large Language Model (LLM) based on LBSN messages.

Keywords: Sentence embeddings, Natural language processing, Location-based social networks, Real-estate marketplace

References

Aiello, L. M., Schifanella, R., Quercia, D., and Aletta, F. (2016). Chatty maps: constructing sound maps of urban areas from social media data. Open Science, 3(3):150690.

Almeida, V. G., Silva, T. R., and Silva, F. A. (2022). Se for, vá na paz: Construindo rotas seguras para veículos coletivos urbanos. In Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos (SBRC), pages 140–153. SBC.

Dubey, A., Naik, N., Parikh, D., Raskar, R., and Hidalgo, C. A. (2016). Deep learning the city: Quantifying urban perception at a global scale. In Proc. of ECCV, pages 196–212. Springer.

Krohling, B., Comarela, G., and Mota, V. F. (2024). Uma arquitetura baseada em redes neurais recorrentes para prediçao de trajetórias veiculares em ambientes urbanos. In Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos (SBRC), pages 379–392. SBC.

Ladeira, L. Z., Santos, F., Cléopas, L., Buteneers, P., and Villas, L. (2022). Neo-nda: Neo natural language data augmentation. In Proc. of ICSC, pages 99–102. IEEE.

Quercia, D., Schifanella, R., Aiello, L. M., Kate, M., et al. (2015). Smelly maps: The digital life of urban smellscapes. In Proc. of ICWSM, pages 327–336, Oxford, UK. AAAI Press.

Rodrigues, D. O., Santos, F. A., Rocha Filho, G. P., Akabane, A. T., Cabral, R., Immich, R., Junior, W. L., Cunha, F. D., Guidoni, D. L., Silva, T. H., et al. (2019). Computação urbana da teoria à prática: Fundamentos, aplicações e desafios. JAI-SBC.

Santos, F. A., Akabane, A. T., Yokoyama, R. S., Loureiro, A. A., and Villas, L. A. (2016). A roadside unit-based localization scheme to improve positioning for vehicular networks. In Proc of VTC-Fall, pages 1–5. IEEE.

Santos, F. A., Kobellarz, J. K., de Souza, F. R., Villas, L. A., and Silva, T. H. (2022). Processamento de linguagem natural em textos de mídias sociais: Fundamentos, ferramentas e aplicações. Sociedade Brasileira de Computação.

Santos, F. A., Rodrigues, D. O., Silva, T. H., Loureiro, A. A., Pazzi, R. W., and Villas, L. A. (2018a). Context-aware vehicle route recommendation platform: Exploring open and crowdsourced data. In Proc. of ICC, pages 1–7. IEEE.

Santos, F. A., Rodrigues, D. O., Silva, T. H., Loureiro, A. A., and Villas, L. A. (2017a). Rotas veiculares cientes de contexto: Arcabouço e aná lise usando dados oficiais e sensoriados por usuários sobre crimes. In Proc. of WGRS. SBC.

Santos, F. A., Silva, T. H., Braun, T., Loureiro, A. A., and Villas, L. A. (2017b). Towards a sustainable people-centric sensing. In Prof. of ICC, pages 1–6. IEEE.

Santos, F. A., Silva, T. H., Loureiro, A. A., Boukerche, A., and Villas, L. A. (2018b). Identificação da reputação de áreas urbanas externas com dados de mídias sociais. In Proc. of SBRC, pages 810–823. SBC.

Santos, F. A., Silva, T. H., Loureiro, A. A., and Villas, L. A. (2020). Automatic extraction of urban outdoor perception from geolocated free texts. Social Network Analysis and Mining, 10:1–23.

Santos, F. A., Silva, T. H., Loureiro, A. A. F., and Villas, L. A. (2018c). Uncovering the perception of urban outdoor areas expressed in social media. In 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI), pages 120–127. IEEE.

Santos, F. A., Silva, T. H., and Villas, L. A. (2024). Real-up: Urban perceptions from lbsns helping moving real-estate market to the next level. In Companion Proceedings of the ACM on Web Conference 2024, pages 1071–1074.