Measuring the semantic similarity of network-time prisms using temporal signatures

Authors: Young Jaegal*, The Ohio State University, Harvey J. Miller, The Ohio State University
Topics: Transportation Geography
Keywords: Time geography, Accessibility, Mobility
Session Type: Paper
Day: 4/3/2019
Start / End Time: 12:40 PM / 2:20 PM
Room: Congressional B, Omni, West
Presentation File: No File Uploaded

The network time prism (NTP) is a time-geographic concept capturing spatiotemporal limits on moving objects within transportation network. Researchers may want to measure the degree of resemblance or similarity among NTPs to sort, cluster or aggregate multiple NTPs, or discover similar patterns of accessibility or exposure for moving objects within transportation networks. Existing NTP similarity measures assess the degree of resemblance based on network topological properties. However, semantic properties relating to attributes distributed within a transportation network can also play a key role in revealing the moving object’s accessibility or exposure to an environment and its resources. Semantic similarity measures for NTPs requires scalable methods to summarizing spatial distribution of the attributes in prisms constrained by transportation network. The objective of this research is to propose a new semantic similarity measure of NTPs. The general strategy is the temporal signature approach that has proven to be effective in measuring similarity of planar STPs: convert a 3D NTP to a time series based on a selected property of the NTP subnetworks at moments in time. Spatial analysis methods developed for network events, such as network autocorrelation, can be used in this approach to measure the semantic properties of NTPs. To demonstrate the proposed method, we perform a case study using a bike-share trip dataset in Columbus, Ohio. The case study constructs NTPs of bike trips taken between bike-share stations and compares NTPs based on the level of bikeability.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login