Volume XLII-2/W7
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W7, 175-179, 2017
https://doi.org/10.5194/isprs-archives-XLII-2-W7-175-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W7, 175-179, 2017
https://doi.org/10.5194/isprs-archives-XLII-2-W7-175-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.

  12 Sep 2017

12 Sep 2017

A WEB-BASED INTERACTIVE PLATFORM FOR CO-CLUSTERING SPATIO-TEMPORAL DATA

X. Wu1, A. Poorthuis1, R. Zurita-Milla2, and M.-J. Kraak2 X. Wu et al.
  • 1Humanities, Arts and Social Sciences, Singapore University of Technology and Design, 8 Somapah Road Singapore 487372, Singapore
  • 2Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente, PO Box 217, 7500 AE, Enschede, the Netherlands

Keywords: Interactive Platform, Web-based, Multiple Linked Visualization, Co-clustering, Spatio-temporal Data

Abstract. Since current studies on clustering analysis mainly focus on exploring spatial or temporal patterns separately, a co-clustering algorithm is utilized in this study to enable the concurrent analysis of spatio-temporal patterns. To allow users to adopt and adapt the algorithm for their own analysis, it is integrated within the server side of an interactive web-based platform. The client side of the platform, running within any modern browser, is a graphical user interface (GUI) with multiple linked visualizations that facilitates the understanding, exploration and interpretation of the raw dataset and co-clustering results. Users can also upload their own datasets and adjust clustering parameters within the platform. To illustrate the use of this platform, an annual temperature dataset from 28 weather stations over 20 years in the Netherlands is used. After the dataset is loaded, it is visualized in a set of linked visualizations: a geographical map, a timeline and a heatmap. This aids the user in understanding the nature of their dataset and the appropriate selection of co-clustering parameters. Once the dataset is processed by the co-clustering algorithm, the results are visualized in the small multiples, a heatmap and a timeline to provide various views for better understanding and also further interpretation. Since the visualization and analysis are integrated in a seamless platform, the user can explore different sets of co-clustering parameters and instantly view the results in order to do iterative, exploratory data analysis. As such, this interactive web-based platform allows users to analyze spatio-temporal data using the co-clustering method and also helps the understanding of the results using multiple linked visualizations.