Line Perception in Parallel Coordinates under different Aspect Ratios

Information

• Publication Type: Master Thesis
• Workgroup(s)/Project(s): not specified
• Date: 2024
• Second Supervisor: Johanna Schmidt
• Open Access: yes
• First Supervisor: Eduard Gröller
• Pages: 113
• Keywords: Parallel Coordinates, Aspect Ratio, Perception

Abstract

This thesis investigates the impact of different aspect ratios on the perception of angles and lines in parallel coordinates. Parallel coordinates are a visualization technique for representing multivariate data where each variable is drawn as a parallel axis, and data points are connected by lines across these axes. This method allows for the simultaneous visualization of more than two variables and enables the interpretation of correlation patterns within a given dataset.However, the reliability and accuracy of this interpretation can be significantly influenced by the aspect ratio of the plot. This thesis aims to explore how variations in aspect ratios affect the accuracy and confidence of users in perceiving correlations within parallel coordinates.The methodological approach comprises three components: the development of a web-based visualization tool, a statistical analysis of line and angle parameters, and an empirical user study. The visualization tool enables users to display parallel coordinates in various aspect ratios and analyze the geometric properties of the lines in the plot. The statistical analysis reveals that aspect ratios significantly correlate with the minimum and maximum angles in parallel coordinates, which in turn affects the visual perception and interpretation of the data. These findings are validated through a web-based user study, demonstrating that specific aspect ratios lead to more accurate and reliable correlation estimates. The results underscore considerate usage of flexible aspect ratios to minimize distortion and ensure the reliability of visual data interpretation in parallel coordinates.

Additional Files and Images

Weblinks

• Entry in reposiTUm (TU Wien Publication Database)
• DOI: 10.34726/hss.2024.119265

BibTeX

@mastersthesis{meka-2024-lpi,
  title =      "Line Perception in Parallel Coordinates under different
               Aspect Ratios",
  author =     "Leon Meka",
  year =       "2024",
  abstract =   "This thesis investigates the impact of different aspect
               ratios on the perception of angles and lines in parallel
               coordinates. Parallel coordinates are a visualization
               technique for representing multivariate data where each
               variable is drawn as a parallel axis, and data points are
               connected by lines across these axes. This method allows for
               the simultaneous visualization of more than two variables
               and enables the interpretation of correlation patterns
               within a given dataset.However, the reliability and accuracy
               of this interpretation can be significantly influenced by
               the aspect ratio of the plot. This thesis aims to explore
               how variations in aspect ratios affect the accuracy and
               confidence of users in perceiving correlations within
               parallel coordinates.The methodological approach comprises
               three components: the development of a web-based
               visualization tool, a statistical analysis of line and angle
               parameters, and an empirical user study. The visualization
               tool enables users to display parallel coordinates in
               various aspect ratios and analyze the geometric properties
               of the lines in the plot. The statistical analysis reveals
               that aspect ratios significantly correlate with the minimum
               and maximum angles in parallel coordinates, which in turn
               affects the visual perception and interpretation of the
               data. These findings are validated through a web-based user
               study, demonstrating that specific aspect ratios lead to
               more accurate and reliable correlation estimates. The
               results underscore considerate usage of flexible aspect
               ratios to minimize distortion and ensure the reliability of
               visual data interpretation in parallel coordinates.",
  pages =      "113",
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Research Unit of Computer Graphics, Institute of Visual
               Computing and Human-Centered Technology, Faculty of
               Informatics, TU Wien",
  keywords =   "Parallel Coordinates, Aspect Ratio, Perception",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2024/meka-2024-lpi/",
}