PhD #01 - Transform GeoLife's plt files into GeoJson
The GeoLife dataset (see below) is stored in a plt file, which are essentially trajectories stored in CSV files. While those are not very handy to use. This script transforms those plt files in GeoJSONs. The output format is inspired by the Moves application's storyline format. The output data consists of a GeoJson FeatureCollection which contains points for locations and MultiLineStrings for trajectories between locations. When activity classification data is available the MultiLineString is split by activity type and the according activity labels are provided.
Download the GeoLife data from the website below and unzip the data. Then simply call the index.js file and provide the following parameters (only the first parameter is required).
node index.js source=./DATA/
source (required) : path to input folder (GeoLife download folder)
target (optional) : path to output folder (is created if it does not exist) / default: ./export/
simplify (optional) : path to output folder (is created if it does not exist) / default: 0.00001
recording_gap (optional) : when is a time gap a recording gap / default 5*60 > (5 Minutes)
loc_spaceline_threshold (optional) : spatial threshold to be occupied in a time loc_spacetime_threshold to be accounted as a location / default: 10 (meters)
loc_spacetime_threshold (optional) : temporal threshold to be spend in an area loc_spaceline_threshold to be accounted as a location / default: 15*60 (15 minutes)
Note: This can be heavy on your machine as the transformation makes use of child processes and launches one process per available cpu. In addition the memory is already increased to 4GB, due to the large files. Be aware, that if reducing simplify files get bigger and the memory might need to be further increased.
The above script will create one geojson file per subject. In addition a summary.json is created providing an overview over the transformed data.
If you want to quickly check if the transformed trajectories are to your satisfaction. Use the vis.html file. Please note that you need to start it from your localhost, otherwise, loading the geojson files will not work.
From April 2007 to August 2012 Microsoft Research Asia's GeoLife project collected trajectories from 182 users. In addition, some of the trajectories contain activity classification data. The project was led by senior research manager Yu Zheng.
The raw data can be downloaded here: https://www.microsoft.com/en-us/download/details.aspx?id=52367
Publications by the GeoLife team can be found here:
[1] Yu Zheng, Lizhu Zhang, Xing Xie, Wei-Ying Ma. Mining interesting locations and travel sequences from GPS trajectories. In Proceedings of International conference on World Wild Web (WWW 2009), Madrid Spain. ACM Press: 791-800.
[2] Yu Zheng, Quannan Li, Yukun Chen, Xing Xie, Wei-Ying Ma. Understanding Mobility Based on GPS Data. In Proceedings of ACM conference on Ubiquitous Computing (UbiComp 2008), Seoul, Korea. ACM Press: 312-321.
[3] Yu Zheng, Xing Xie, Wei-Ying Ma, GeoLife: A Collaborative Social Networking Service among User, location and trajectory. Invited paper, in IEEE Data Engineering Bulletin. 33, 2, 2010, pp. 32-40.
All code besides the datasets in the folder test is available under GPLv3. The data sets provided for testing are part of the GeoLife dataset. The GeoLife dataset does not clearly state under what license it is provided. The three exemplary datasets are rather small so I hope its okey.