Smart Brushing for Parallel Coordinates

Abstract

The Parallel Coordinates plot is a popular tool for the visualization of high-dimensional data. One of the main challenges when using parallel coordinates is occlusion and overplotting resulting from large data sets. Brushing is a popular approach to address these challenges. Since its conception, limited improvements have been made to brushing both in the form of visual design and functional interaction. We present a set of novel, smart brushing techniques that enhance the standard interactive brushing of a parallel coordinates plot. We introduce two new interaction concepts: Higher-order, sketch-based brushing, and smart, data-driven brushing. Higher-order brushes support interactive, flexible, n-dimensional pattern searches involving an arbitrary number of dimensions. Smart, data-driven brushing provides interactive, real-time guidance to the user during the brushing process based on derived meta-data. In addition, we implement a selection of novel enhancements and user options that complement the two techniques as well as enhance the exploration and analytical ability of the user. We demonstrate the utility and evaluate the results using a case study with a large, high-dimensional, real-world telecommunication data set and we report domain expert feedback from the data suppliers.

Visualising Business Data: A Survey

“…Wright [47] Proprietary Case Study from portfolio management, derivatives management, customer credit scores Gresh and Kelton [78] Proprietary Private IBM business by-product data Eick [79] Proprietary Log data from web servers used to analyse the efficiency of their website Burkhard [23] Proprietary Case study from Swiss Federal Institute of Technology using business strategy data Vliegen et al [48] Proprietary Unspecified business data Keim et al [80] Proprietary Transaction datasets Otsuka et al [20] Proprietary Digital nametags collect employee interaction data Sedlmair et al [24] Survey Existing software evaluation Kandel et al [18] Proprietary Interview Study with industry experts Du et al [70] Survey A survey of business process visualisation literature Aigner [25] Proprietary Text from interview study Broeksema et al [71] Proprietary Decision model data Bai et al [49] Proprietary Geospatial data for utility network coverage Lafon et al [26] Proprietary User Study of unspecified business data visualisation Nicholas et al [50] Proprietary Private customer survey database from automotive company Roberts et al [51] Proprietary Private call centre interaction database Ghooshchi et al [72] Proprietary Business Processes from undefined source Kumar and Belwal [52] Public Multiple public data sources looking at different aspects of a business Bachhofner et al [73] Proprietary Business processes from industry contacts Lea et al [74] Proprietary Business process data was used alongside simulated data to test prototypes Roberts et al [53] Proprietary Call centre event data from industry partner…”

“…The secondary classification used in the survey is shown in Table 1. Aigner [25] Broeksema et al [71] Bai et al [49] Lafon et al [26] Ferreira et al [54] Basole et al [58] Hao et al [42] 2014 Nicholas et al [50] Basole [76] Deligiannidis and Noyes [59] Basole and Bellamy [77] Lu et al [35] Shi et al [36] Rodden [84] Yaeli et al [21] Saitoh [43] 2015 Keahey [29] Basole et al [60] Dou et al [32] Kameoka et al [66] Nair et al [85] 2016 Roberts et al [51] Liu et al [81] Iyer and Basole [61] Basole et al [31] Wu et al [67] Nagaoka et al [22] Sijtsma et al [37] 2017 Ghooshchi et al [72] Kumar and Belwal [52] Bachhofner et al [73] Ramesh et al [34] Schotter et al [62] Kang et al [69] Fayoumi et al [44] 2018 Lea et al [74] Roberts et al [53] Basole et al [63] Sathiyanarayanan et al [68] Haleem et al [45] Saga and Yagi [46] Primary Data as Intentional, Active Digital Collection (CB): The following research conveys geo-location data collected through hardware that is used to track customer be...…”

“…Among the most recently published of the literature, both Roberts et al [51,53] and Wu et al [67] visualised telecommunications data. The future work discussions in each of these papers emphasise the need for more features to be added to their software in an attempt to extract all the value from the data.…”

“…The sketch brush placement with the dynamic brush interval range glyphs which were automatically placed at 1 standard deviation from the mean. Image courtesy of Roberts et al[53].…”

Feature-Rich, GPU-Assisted Scatterplots for Millions of Call Events

“…Roberts et al present an interactive treemap application for displaying call metrics of calls serviced at a call center over one day [30]. Roberts et al also use the same data to demonstrate a higher-order brushing technique for parallel co-ordinate plots [31]. Their data set is limited to one day only, while this work can render a complete month's worth of data.…”