Visualizing the Impact of Geographical Variations on Multivariate Clustering

Abstract

Traditional multivariate clustering approaches are common in many geovisualization applications. These algorithms are used to define geodemographic profiles, ecosystems and various other land use patterns that are based on multivariate measures. Cluster labels are then projected onto a choropleth map to enable analysts to explore spatial dependencies and heterogeneity within the multivariate attributes. However, local variations in the data and choices of clustering parameters can greatly impact the resultant visualization. In this work, we develop a visual analytics framework for exploring and comparing the impact of geographical variations for multivariate clustering. Our framework employs a variety of graphical configurations and summary statistics to explore the spatial extents of clustering. It also allows users to discover patterns that can be concealed by traditional global clustering via several interactive visualization techniques including a novel drag & drop clustering difference view. We demonstrate the applicability of our framework over a demographics dataset containing quick facts about counties in the continental United States and demonstrate the need for analytical tools that can enable users to explore and compare clustering results over varying geographical features and scales.

The State‐of‐the‐Art in Predictive Visual Analytics

“…Pilhöfer et al [PGU12] use Bertin's Classification Criterion to optimize the display order of nominal variables and better interpret and compare clustering results from different models. Other techniques have explored methods for visually comparing clustering results under different parameters [MMT*14, ZLMM16, ZM17] in geographical displays.…”

Commercial Visual Analytics Systems–Advances in the Big Data Analytics Field

“…Cao et al present a treemap-like glyph for a comparative analysis of multidimensional cluster results [41]. Some recent works focus on cluster comparison and ensemble building [42], [43]. More generally, Sacha et al claim that a (visual) comparison of data models can help in the knowledge generation process [44].…”

“…However, most VA research prototypes rely on advanced drill-down functionality [73]- [75]. Multi-scale analysis, such as presented for geo-related data [76] or text data [43], view distortion and adoption techniques, such as presented for Treemaps by Tu and Shen [77], as well as novel navigation concepts, such as link-sliding [78], are only found in research prototypes.…”

Recent research advances on interactive machine learning

“…Thus, visual cluster analysis has been applied to many data-driven applications such as weather ensemble forecast (Kumpf et al, 2018) and air traffic optimization (Andrienko et al, 2018). Recent work mainly focuses on cluster-based exploratory data analysis (Andrienko et al, 2018;Badam et al, 2017;Heimerl et al, 2016;Sacha et al, 2018;Wu et al, 2017b,c), comparative clustering analysis (Jarema et al, 2015;Kumpf et al, 2018;Kwon et al, 2018;Zhang et al, 2016b), and bi-cluster analysis (Sun et al, 2016;Watanabe et al, 2015;Wu et al, 2017aWu et al, , 2015Zhao et al, 2018).…”