Analysis of WiSARD-Based Models for Bus Trajectory Classification in the Context of Urban Mobility

• Bruno O. Barbosa Federal Center for Technological Education Celso Suckow da Fonseca (CEFET/RJ)
• Lucas A. B. Marinho Federal Center for Technological Education Celso Suckow da Fonseca (CEFET/RJ)
• Marcelo D. Santos Federal Center for Technological Education Celso Suckow da Fonseca (CEFET/RJ)
• Fernando Alexandrino Federal Institute of Education, Science, and Technology of São Paulo (IFSP) / Federal Center for Technological Education Celso Suckow da Fonseca (CEFET/RJ)
• Rafaelli Coutinho Federal Center for Technological Education Celso Suckow da Fonseca (CEFET/RJ)
• Ubiratam de Paula Rural Federal University of Rio de Janeiro (UFRRJ)
• Diego Carvalho Federal Center for Technological Education Celso Suckow da Fonseca (CEFET/RJ)
• Eduardo Ogasawara Federal Center for Technological Education Celso Suckow da Fonseca (CEFET/RJ)

DOI: https://doi.org/10.5753/bresci.2024.243296

Abstract

Road infrastructure is essential in the urban context, and the increase in automobile use negatively impacts the quality of life due to heavy traffic congestion. Analyzing vehicle trajectories is crucial for the efficient management of cities. By monitoring the GPS of these vehicles, it is possible to classify trajectories and detect deviations, among other analyses. Weightless neural networks (WiSARD) have recently been used for these tasks. However, the performance of these networks is influenced by the encodings adopted in their inputs. This work evaluates the task of trajectory classification for the road transport system of the municipality of Rio de Janeiro to study this influence. The results show significant efficiency in the analyses, with subtle differences in the applicability of each representation, determining the efficiency of each based on classification quality and execution time.

Keywords: WiSARD, Encoding, Trajectory Classification, Bus, Urban mobility

References

Aleksander, I., Thomas, W., and Bowden, P. (1984). Wisard a radical step forward in image recognition. Sensor Review, 4(3):120 – rere.

Barbosa, R., Cardoso, D. O., Carvalho, D., and França, F. M. (2018). Weightless neurosymbolic GPS trajectory classification. Neurocomputing, 298:100 – 108.

Bentley, J. L. (1975). Multidimensional Binary Search Trees Used for Associative Searching. Communications of the ACM, 18(9):509 – 517.

Carneiro, C. (2012). A Função do Índice de Síntese das Linguagens na Classificação Gramatical com Redes Neurais sem Peso. Mestrado em Engenharia de Sistemas e Computação, Universidade Federal do Rio de Janeiro, Rio de Janeiro.

Chen, W., Guo, F., andWang, F.-Y. (2015). A Survey of Traffic Data Visualization. IEEE Transactions on Intelligent Transportation Systems, 16(6):2970 – 2984.

De Souza, D. F. P., França, F. M. G., and Lima, P. M. V. (2014). Spatio-temporal pattern classification with KernelCanvas and WiSARD. In Proceedings - 2014 Brazilian Conference on Intelligent Systems, BRACIS 2014, pages 228 – 233.

Instituto Pereira Passos (2017). Data.Rio - DataLake. Technical report, [link].

Lee, J., Han, J., Li, X., and Gonzalez, H. (2008). TraClass: Trajectory classification using hierarchical region based and trajectory based clustering. In Proceedings of the VLDB Endowment, volume 1, pages 1081 – 1094.

Pontes Cotta, K., Sena Ferreira, R., and Franca, F. M. (2018). Weightless neural network WiSARD applied to online recommender systems. In Proceedings - 2018 Brazilian Conference on Intelligent Systems, BRACIS 2018, pages 348 – 353.

T. L. Bacellar, A., Susskind, Z., A. Q. Villon, L., D. S. Miranda, I., Santiago De Araújo, L., Cadette Dutra, D. L., Breternitz Jr., M., John, L., Lima, P., and França, F. (2022). Distributive Thermometer: A New Unary Encoding for Weightless Neural Networks. In ESANN 2022 proceedings, pages 31–36, Bruges (Belgium) and online event. Ciaco - [link].

Valença, G. and Santos, E. (2020). The relationship between the concept of complete streets and the National Urban Mobility Policy: Application to a road project in Natal-RN, Brazil. Eure, 46(139):73 – 89.