Journal cover Journal topic
Proceedings of the ICA
Journal topic
Articles | Volume 1
https://doi.org/10.5194/ica-proc-1-29-2018
https://doi.org/10.5194/ica-proc-1-29-2018
16 May 2018
 | 16 May 2018

Interactive visual exploration and analysis of origin-destination data

Linfang Ding, Liqiu Meng, Jian Yang, and Jukka M. Krisp

Keywords: Origin-destination, Interactive clustering, Parallel coordinates, Gradient line rendering

Abstract. In this paper, we propose a visual analytics approach for the exploration of spatiotemporal interaction patterns of massive origin-destination data. Firstly, we visually query the movement database for data at certain time windows. Secondly, we conduct interactive clustering to allow the users to select input variables/features (e.g., origins, destinations, distance, and duration) and to adjust clustering parameters (e.g. distance threshold). The agglomerative hierarchical clustering method is applied for the multivariate clustering of the origin-destination data. Thirdly, we design a parallel coordinates plot for visualizing the precomputed clusters and for further exploration of interesting clusters. Finally, we propose a gradient line rendering technique to show the spatial and directional distribution of origin-destination clusters on a map view. We implement the visual analytics approach in a web-based interactive environment and apply it to real-world floating car data from Shanghai. The experiment results show the origin/destination hotspots and their spatial interaction patterns. They also demonstrate the effectiveness of our proposed approach.

Publications Copernicus
Download
Citation