Classificação do Modo de Transporte e Propósito de Viagem Baseada em Dados Socioeconômicos
Resumo
Características socioeconômicas, como renda e idade, podem influenciar os padrões de mobilidade observados nos residentes de uma cidade que por sua vez podem ser restringidos pelas opções e qualidade dos meios de transporte disponibilizados. Dado isso, uma das principais tarefas durante a caracterização de dados de mobilidade é a identificação dos meios de transporte utilizados mais frequentemente durante o deslocamento dos cidadãos, assim como dos propósitos de suas viagens. Este trabalho, propõe uma técnica para identificação dos meios de transporte utilizados em uma viagem assim como o seu propósito, baseando-se somente em variáveis socioeconômicas. O método proposto é avaliado através de experimentos de validação cruzada utilizando uma base de dados pública da cidade de Nova Iorque. Através da comparação das matrizes de confusão da obtidas com diferentes algoritmos de aprendizado supervisionado é possível concluir que o método proposto apresenta um desempenho similar as demais técnicas na classificação do propósito da viagem, com acurácia de 67%, e superior para a classificação do modo de transporte, com acurácia de 82%.
Palavras-chave:
Análise e Mineração de Dados, Mobilidade Humana, Sistemas de Transporte Inteligentes
Referências
Bedogni, L., Di Felice, M., and Bononi, L. (2016). Context-aware android applicati-ons through transportation mode detection techniques. Wireless Communications andMobile Computing, 16(16).
Bohte, W. and Maat, K. (2009). Deriving and validating trip purposes and travel modesfor multi-day gps-based travel surveys: A large-scale application in the netherlands.Transportation Research Part C: Emerging Technologies, 17(3):285-297.
Breiman, L. (2001). Random forests. Machine learning, 45(1):5-32.
Calabrese, F., Ferrari, L., and Blondel, V. D. (2014). Urban sensing using mobile phonenetwork data: a survey of research. ACM Computing Surveys, 47(2):1-20.
Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote:synthetic minority over-sampling technique. Journal of artificial intelligence research,16:321-357.
Ermagun, A., Fan, Y., Wolfson, J., Adomavicius, G., and Das, K. (2017). Real-time trippurpose prediction using online location-based search and discovery services. Trans-portation Research Part C: Emerging Technologies, 77:96-112.
Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters,27(8):861 — 874.
Geurs, K. T., Thomas, T., Bijlsma, M., and Douhou, S. (2015). Automatic trip and modedetection with move smarter: First results from the dutch mobile mobility panel. Trans-portation research procedia, 11:247-262.
Gong, L., Kanamori, R., and Yamamoto, T. (2017). Data selection in machine learningfor identifying trip purposes and travel modes from longitudinal gps data collectionlasting for seasons. Travel Behaviour and Society.
Ma, Y. and Guo, G. (2014). Support vector machines applications. Springer.
Mäenpää, H., Lobov, A., and Lastra, J. L. M. (2017). Travel mode estimation for multi-modal journey planner. Transportation Research Part C: Emerging Technologies,82:273-289.
Meng, C., Cui, Y., He, Q., Su, L., and Gao, J. (2017). Travel purpose inference with gpstrajectories, pois, and geo-tagged social media data. In IEEE International Conferenceon Big Data (BigData 2017), Boston, USA, pages 1319-1324.
Montini, L., Prost, S., Schrammel, J., Rieser-Schiissler, N., and Axhausen, K. W. (2015).Comparison of travel diaries generated from smartphone data and dedicated gps devi-ces. Transportation Research Procedia, 11:227-241.
Montini, L., Rieser-Schiissler, N., and Axhausen, K. W. (2014a). Personalisation in multi-day gps and accelerometer data processing. In 14th Swiss Transport Research Confe-rence (STRC), Ascona, Switzerland.
Montini, L., Rieser-Schiissler, N., Horni, A., and Axhausen, K. (2014b). Trip purposeidentification from gps tracks. Transportation Research Record: Journal of the Trans-portation Research Board, (2405):16-23.
Morris, S., Humphrey, A., Pickering, A., Tipping, S., Templeton, I., and Hurn, J. (2013).National travel survey 2013. The Department for Transport, NatCen Social Research,London, UK.
NYC Department of Transportation (2017). Citywide mobility survey. Data re-treved from NYC Open Data, https://data.cityofnewyork.us/Transportation/Citywide-Mobility-Survey-Trip-Diary/mpk5-48av.
Quintella, C. A.M. S., Andrade, L. C., and Campos, C. A. V. (2016). Detecting the trans-portation mode for context-aware systems using smartphones. In IEEE Conference onon Intelligent Transportation Systems (ITSC'2016), Rio de Janeiro, Brazil.
Schrammel, J., Busch, M., and Tscheligi, M. (2013). Peacox-persuasive advisor for co2-reducing cross-modal trip planning. In 8th International Conference on PersuasiveTechnology (PERSUASIVE), Sidney, Australia.
Seo, T., Kusakabe, T., Gotoh, H., and Asakura, Y. (2017). Interactive online machinelearning approach for activity-travel survey. Transportation Research Part B: Metho-dological.
Soares, E. F. d. S., Revoredo, K., Baião, F., de MS Quintella, C. A., and Campos, C. A. V.(2019). A combined solution for real-time travel mode detection and trip purposeprediction. IEEE Transactions on Intelligent Transportation Systems, 20(12):4655-4664.
Soares, E. F. S., Quintella, C. A. M. S., and Campos, C. A. V. (2017). Towards anapplication for real-time travel mode detection in urban centers. In IEEE VehicularTechnology Conference (VTC-Fall'2017), Toronto, Canada.
Song, Y.-Y. and Ying, L. (2015). Decision tree methods: applications for classificationand prediction. Shanghai archives of psychiatry, 27(2):130.
Sperandei, S. (2014). Understanding logistic regression analysis. Biochemia medica:Biochemia medica, 24(1):12-18.
Stopher, P. R., Jiang, Q., FitzGerald, C., et al. (2005). Processing gps data from travelsurveys. In 2nd international collogium on the behavioural foundations of integratedland-use and transportation models: frameworks, models and applications.
Van der Vaart, A. W. (2000). Asymptotic statistics, volume 3. Cambridge university press.
Wang, B., Gao, L., and Juan, Z. (2017a). Travel mode detection using gps data andsocioeconomic attributes based on a random forest classifier. IEEE Transactions onIntelligent Transportation Systems.
Wang, H., Liu, G., Duan, J., and Zhang, L. (2017b). Detecting transportation modes usingdeep neural network. JEICE Transactions on Information and Systems, 100(5):1132-1135.
Xiao, G., Juan, Z., and Zhang, C. (2016). Detecting trip purposes from smartphone-based travel surveys with artificial neural networks and particle swarm optimization.Transportation Research Part C: Emerging Technologies, 71:447-463.
Ye, N., Gao, L., Juan, Z., and Ni, A. (2018). Are people from households with chil-dren more likely to travel by car? an empirical investigation of individual travel modechoices in shanghai, china. Sustainability, 10(12):4573.
Zhang, J., Wang, F.-Y., Wang, K., Lin, W.-H., Xu, X., and Chen, C. (2011). Data-drivenintelligent transportation systems: A survey. IEEE Transactions on Intelligent Trans-portation Systems, 12(4):1624-1639.
Bohte, W. and Maat, K. (2009). Deriving and validating trip purposes and travel modesfor multi-day gps-based travel surveys: A large-scale application in the netherlands.Transportation Research Part C: Emerging Technologies, 17(3):285-297.
Breiman, L. (2001). Random forests. Machine learning, 45(1):5-32.
Calabrese, F., Ferrari, L., and Blondel, V. D. (2014). Urban sensing using mobile phonenetwork data: a survey of research. ACM Computing Surveys, 47(2):1-20.
Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote:synthetic minority over-sampling technique. Journal of artificial intelligence research,16:321-357.
Ermagun, A., Fan, Y., Wolfson, J., Adomavicius, G., and Das, K. (2017). Real-time trippurpose prediction using online location-based search and discovery services. Trans-portation Research Part C: Emerging Technologies, 77:96-112.
Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters,27(8):861 — 874.
Geurs, K. T., Thomas, T., Bijlsma, M., and Douhou, S. (2015). Automatic trip and modedetection with move smarter: First results from the dutch mobile mobility panel. Trans-portation research procedia, 11:247-262.
Gong, L., Kanamori, R., and Yamamoto, T. (2017). Data selection in machine learningfor identifying trip purposes and travel modes from longitudinal gps data collectionlasting for seasons. Travel Behaviour and Society.
Ma, Y. and Guo, G. (2014). Support vector machines applications. Springer.
Mäenpää, H., Lobov, A., and Lastra, J. L. M. (2017). Travel mode estimation for multi-modal journey planner. Transportation Research Part C: Emerging Technologies,82:273-289.
Meng, C., Cui, Y., He, Q., Su, L., and Gao, J. (2017). Travel purpose inference with gpstrajectories, pois, and geo-tagged social media data. In IEEE International Conferenceon Big Data (BigData 2017), Boston, USA, pages 1319-1324.
Montini, L., Prost, S., Schrammel, J., Rieser-Schiissler, N., and Axhausen, K. W. (2015).Comparison of travel diaries generated from smartphone data and dedicated gps devi-ces. Transportation Research Procedia, 11:227-241.
Montini, L., Rieser-Schiissler, N., and Axhausen, K. W. (2014a). Personalisation in multi-day gps and accelerometer data processing. In 14th Swiss Transport Research Confe-rence (STRC), Ascona, Switzerland.
Montini, L., Rieser-Schiissler, N., Horni, A., and Axhausen, K. (2014b). Trip purposeidentification from gps tracks. Transportation Research Record: Journal of the Trans-portation Research Board, (2405):16-23.
Morris, S., Humphrey, A., Pickering, A., Tipping, S., Templeton, I., and Hurn, J. (2013).National travel survey 2013. The Department for Transport, NatCen Social Research,London, UK.
NYC Department of Transportation (2017). Citywide mobility survey. Data re-treved from NYC Open Data, https://data.cityofnewyork.us/Transportation/Citywide-Mobility-Survey-Trip-Diary/mpk5-48av.
Quintella, C. A.M. S., Andrade, L. C., and Campos, C. A. V. (2016). Detecting the trans-portation mode for context-aware systems using smartphones. In IEEE Conference onon Intelligent Transportation Systems (ITSC'2016), Rio de Janeiro, Brazil.
Schrammel, J., Busch, M., and Tscheligi, M. (2013). Peacox-persuasive advisor for co2-reducing cross-modal trip planning. In 8th International Conference on PersuasiveTechnology (PERSUASIVE), Sidney, Australia.
Seo, T., Kusakabe, T., Gotoh, H., and Asakura, Y. (2017). Interactive online machinelearning approach for activity-travel survey. Transportation Research Part B: Metho-dological.
Soares, E. F. d. S., Revoredo, K., Baião, F., de MS Quintella, C. A., and Campos, C. A. V.(2019). A combined solution for real-time travel mode detection and trip purposeprediction. IEEE Transactions on Intelligent Transportation Systems, 20(12):4655-4664.
Soares, E. F. S., Quintella, C. A. M. S., and Campos, C. A. V. (2017). Towards anapplication for real-time travel mode detection in urban centers. In IEEE VehicularTechnology Conference (VTC-Fall'2017), Toronto, Canada.
Song, Y.-Y. and Ying, L. (2015). Decision tree methods: applications for classificationand prediction. Shanghai archives of psychiatry, 27(2):130.
Sperandei, S. (2014). Understanding logistic regression analysis. Biochemia medica:Biochemia medica, 24(1):12-18.
Stopher, P. R., Jiang, Q., FitzGerald, C., et al. (2005). Processing gps data from travelsurveys. In 2nd international collogium on the behavioural foundations of integratedland-use and transportation models: frameworks, models and applications.
Van der Vaart, A. W. (2000). Asymptotic statistics, volume 3. Cambridge university press.
Wang, B., Gao, L., and Juan, Z. (2017a). Travel mode detection using gps data andsocioeconomic attributes based on a random forest classifier. IEEE Transactions onIntelligent Transportation Systems.
Wang, H., Liu, G., Duan, J., and Zhang, L. (2017b). Detecting transportation modes usingdeep neural network. JEICE Transactions on Information and Systems, 100(5):1132-1135.
Xiao, G., Juan, Z., and Zhang, C. (2016). Detecting trip purposes from smartphone-based travel surveys with artificial neural networks and particle swarm optimization.Transportation Research Part C: Emerging Technologies, 71:447-463.
Ye, N., Gao, L., Juan, Z., and Ni, A. (2018). Are people from households with chil-dren more likely to travel by car? an empirical investigation of individual travel modechoices in shanghai, china. Sustainability, 10(12):4573.
Zhang, J., Wang, F.-Y., Wang, K., Lin, W.-H., Xu, X., and Chen, C. (2011). Data-drivenintelligent transportation systems: A survey. IEEE Transactions on Intelligent Trans-portation Systems, 12(4):1624-1639.
Publicado
10/12/2020
Como Citar
FIGUEIREDO DE SOUZA SOARES, Elton; VIEIRA CAMPOS, Carlos Alberto.
Classificação do Modo de Transporte e Propósito de Viagem Baseada em Dados Socioeconômicos. In: WORKSHOP DE COMPUTAÇÃO URBANA (COURB), 4. , 2020, Rio de Janeiro.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2020
.
p. 220-233.
ISSN 2595-2706.
DOI: https://doi.org/10.5753/courb.2020.12365.
