{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/inferring-transportation-modes-from-gps","title":"Inferring transportation modes from GPS trajectories using a convolutional neural network","arxiv_id":"1804.02386","date":"2018-04-05","proceeding":null,"authors":["Sina Dabiri","Kevin Heaslip"],"abstract":"Identifying the distribution of users' transportation modes is an essential\npart of travel demand analysis and transportation planning. With the advent of\nubiquitous GPS-enabled devices (e.g., a smartphone), a cost-effective approach\nfor inferring commuters' mobility mode(s) is to leverage their GPS\ntrajectories. A majority of studies have proposed mode inference models based\non hand-crafted features and traditional machine learning algorithms. However,\nmanual features engender some major drawbacks including vulnerability to\ntraffic and environmental conditions as well as possessing human's bias in\ncreating efficient features. One way to overcome these issues is by utilizing\nConvolutional Neural Network (CNN) schemes that are capable of automatically\ndriving high-level features from the raw input. Accordingly, in this paper, we\ntake advantage of CNN architectures so as to predict travel modes based on only\nraw GPS trajectories, where the modes are labeled as walk, bike, bus, driving,\nand train. Our key contribution is designing the layout of the CNN's input\nlayer in such a way that not only is adaptable with the CNN schemes but\nrepresents fundamental motion characteristics of a moving object including\nspeed, acceleration, jerk, and bearing rate. Furthermore, we ameliorate the\nquality of GPS logs through several data preprocessing steps. Using the clean\ninput layer, a variety of CNN configurations are evaluated to achieve the best\nCNN architecture. The highest accuracy of 84.8% has been achieved through the\nensemble of the best CNN configuration. In this research, we contrast our\nmethodology with traditional machine learning algorithms as well as the seminal\nand most related studies to demonstrate the superiority of our framework.","url_abs":"http://arxiv.org/abs/1804.02386v1","url_pdf":"http://arxiv.org/pdf/1804.02386v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"inferring-transportation-modes-from-gps","repo_url":"https://github.com/PatrickMotylinski/LBCPI-project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}