Presentation information

International Session

International Session » [ES] E-2 Machine learning

[4D3-E-2] Machine learning: living environment

Fri. Jun 7, 2019 2:00 PM - 3:40 PM Room D (301B Medium meeting room)

Chair: Junichiro Mori (The University of Tokyo)

2:40 PM - 3:00 PM

[4D3-E-2-03] Scoring and Classifying Regions via Multimodal Transportation Networks

〇Aaron Bramson1,2,3,4, Megumi Hori1, Zha Bingran1, Hirohisa Inamoto1 (1. GA Technologies, 2. RIKEN Center for Biosystems Dynamics Research, 3. Ghent Univeristy, 4. UNC - Charlotte)

Keywords:Transportation Networks, Accessibility, Clustering, Network Similarity

In order to better understand the role of transportation convenience in location preferences, as well as to uncover transportation system patterns that span multiple modes of transportation, we score geographic regions according to properties of their multimodal transportation networks. The various scores are then used to classify regions by their dominant mode of transportation, and rank/cluster regions by their transportation features. Specifically, we analyze the train, bus, and road networks of major cities and neighborhoods of Japan to classify them as being train-centric, bus-centric, or car-centric. We also generate scores based on various transportation features to rank cities by their access to public transportation and to categorize/cluster neighborhoods of major cities by their transportation and accessibility properties. We find that business hubs (having low populations) are conveniently reachable via public transportation but vary greatly in their automobile accessibility. Suburban regions have lower connectivity overall but are typically strongly connected to at least one business area. As increasingly rural areas rely more strongly on the road and bus networks, but the network features do not correlate highly with population density.