Volume XLII-4/W18
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W18, 717–720, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W18-717-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W18, 717–720, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W18-717-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  18 Oct 2019

18 Oct 2019

EXTRACTING POINT OF INTERESTS FROM MOVEMENT DATA USING KERNEL DENSITY AND WEIGHTED K-MEANS

M. Malekzadeh, R. Javanmard, and F. Karimipour M. Malekzadeh et al.
  • School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Iran

Keywords: Weighted K-means, Trajectory, Movement data, Kernel Density, Point of Interest

Abstract. Development in spatial data acquisition techniques, facilitate the process of analyzing movement characteristics and removed the lack of spatial data challenge. Annually, an enormous amount of spatial data are produced, and interpretation of this volume of data has become a major challenge. In this study, the movement data of 157 users in Geneva, Switzerland, were used and attempted to analyze their movement patterns. After the pre-processing stage, in order to investigate the dense areas, Kernel Density is calculated for each point for its neighborhood. The size of each cell of the output raster is approximately 100 meter. Afterward, in order to find the point of interests in the Geneva city, Weighted K-means is used for clustering of the raster. The kernel value of each cell is considered as the weight of the cell. Subsequently, the centroid of each final cluster has reflected the point of interest. As a final point, with the intention of assessing the results, the land use of the area is compared to each point of interest. Eventually, an interpretation is given.