一種基於蜂窩資料對公共運輸模式偵測與預測的通用框架

A General Framework for Public Transport Mode Detection and Prediction Based on Cellular Data

勾小川(Xiao-Chuan Gou)

指導教授 ： 彭文志

國立交通大學/資訊學院/資訊科學與工程研究所/碩士(2019年)

摘要

人群搭乘公共運輸工具的記錄被廣泛應用于交通規劃，人流預測以及基於位置行銷等諸多方面。通過從人群原始軌跡資料中偵測并預測出這些記錄能夠為這些應用提供很大幫助。與此同時，手機在人們日常生活中被廣泛使用，由此每天都會產生巨量資料。本文提出了一個通用框架，該框架提供了線下偵測和線上預測用戶搭乘公共運輸的功能。這些功能基於用戶的基地台蜂窩資料和公共交通網路信息。我們提出了一個基於格網化的模型來解決偵測任務，同時提供了基於位置過濾器的長短期記憶模型(LSTM)。為了獲得更為精準的偵測結果，我們解決了不同公共運輸網路間及內部的重疊問題。在實驗的部分，我們提出了一種基於Google API的新的評估方法來對我們的偵測結果評估。為了能夠更好地訓練預測模型，我們重構了一個新的訓練集并評估了它的效果。

關鍵字

城市計算 ； 移動模式偵測 ； 蜂窩資料

並列摘要

Users' records of taking public transport are used in traffic planning, crowd flow prediction and location-based marketing. As the same time, mobile phones are used widely in our life and they generate a large amount of data each day. In this paper, we propose a general framework containing two parts: offline detecting and online predicting public transportation trips based on user’s cellular data and urban transport network location data. We propose a grid-based model for detection and a LSTM model with location filter for prediction. To get more precisely result of detection, we resolve the overlapping problem among the detection results from different transport network. In the experiment part, we propose a new method based on Google API to estimate the detection problem results. For the better performance of the prediction model, we rebuild a new training dataset and estimate its effect.

並列關鍵字

City computing ； Transport mode detection ； Cellular data

參考文獻

[1] C. Zhang, H. Wang, and H. Xiong, “An automatic approach for transit advertising in public transportation systems,” in 2017 IEEE International Conference on Data Mining(ICDM), Nov 2017, pp. 1183–1188.

[2] Y. Zheng, L. Liu, L. Wang, and X. Xie, “Learning transportation mode from raw gps data for geographic applications on the web,” in Proceedings of the 17th International Conference on World Wide Web, ser. WWW ’08. New York, NY, USA: ACM, 2008, pp.247–256. [Online]. Available: http://doi.acm.org/10.1145/1367497.1367532

[3] Y. Endo, H. Toda, K. Nishida, and A. Kawanobe, “Deep feature extraction from trajectories for transportation mode estimation,” in Proceedings, Part II, of the 20th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining - Volume 9652, ser. PAKDD 2016. Berlin, Heidelberg: Springer-Verlag, 2016, pp. 54–66. [Online].Available: https://doi.org/10.1007/978-3-319-31750-25

[4] Y. Zheng, Y. Chen, Q. Li, X. Xie, and W.-Y. Ma, “Understanding transportation modes based on gps data for web applications,” ACM Trans. Web, vol. 4, no. 1, pp. 1:1–1:36, Jan. 2010. [Online]. Available: http://doi.acm.org/10.1145/1658373.1658374

[5] Y. Qu, H. Gong, and P. Wang, “Transportation mode split with mobile phone data,” in 2015 IEEE 18th International Conference on Intelligent Transportation Systems, Sept 2015, pp. 285–289.