High Dimensional Clustering Using Parallel Coordinates and the Grand Tour

Wegman, Edward J.; Luo, Qiang

doi:10.1007/978-3-642-59051-1_10

Cited by 82 publications

(91 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Analyzing and visualizing multidimensional data are certainly not a new endeavor, and there have been a number of innovative approaches to this problem (22)(23)(24)(25)(26)(27)(28)(29)(30). Most all these methods are intent on mapping multidimensional populations into two-or three-dimensional space after some type of clustering algorithm.…”

Section: Discussionmentioning

confidence: 99%

Probability state modeling theory

Bagwell¹,

Hunsberger²,

Herbert³

et al. 2015

Cytometry Pt A

View full text Add to dashboard Cite

As the technology of cytometry matures, there is mounting pressure to address two major issues with data analyses. The first issue is to develop new analysis methods for high-dimensional data that can directly reveal and quantify important characteristics associated with complex cellular biology. The other issue is to replace subjective and inaccurate gating with automated methods that objectively define subpopulations and account for population overlap due to measurement uncertainty. Probability state modeling (PSM) is a technique that addresses both of these issues. The theory and important algorithms associated with PSM are presented along with simple examples and general strategies for autonomous analyses. PSM is leveraged to better understand B-cell ontogeny in bone marrow in a companion Cytometry Part B manuscript. Three short relevant videos are available in the online supporting information for both of these papers. PSM avoids the dimensionality barrier normally associated with high-dimensionality modeling by using broadened quantile functions instead of frequency functions to represent the modulation of cellular epitopes as cells differentiate. Since modeling programs ultimately minimize or maximize one or more objective functions, they are particularly amenable to automation and, therefore, represent a viable alternative to subjective and inaccurate gating approaches. V C 2015 International Society for Advancement of Cytometry

show abstract

Section: Discussionmentioning

confidence: 99%

Probability state modeling theory

Bagwell¹,

Hunsberger²,

Herbert³

et al. 2015

Cytometry Pt A

View full text Add to dashboard Cite

show abstract

“…However, such techniques can still be applied on the line representation portion of the SPPC to facilitate data comprehension. Methods such as grand tour [3,38] which examines structure of high dimensional data from all possible angles, and projection pursuit [14] which only shows important aspects of high dimensional space, can be applied to our system. DIMDS creates new configurations starting from a scatterplot of neighboring axis whose Stress value is zero.…”

Section: Discussionmentioning

confidence: 99%

Scattering Points in Parallel Coordinates

Yuan

Guo

et al. 2009

IEEE Trans. Visual. Comput. Graphics

111

View full text Add to dashboard Cite

Abstract-In this paper, we present a novel parallel coordinates design integrated with points (Scattering Points in Parallel Coordinates, SPPC), by taking advantage of both parallel coordinates and scatterplots. Different from most multiple views visualization frameworks involving parallel coordinates where each visualization type occupies an individual window, we convert two selected neighboring coordinate axes into a scatterplot directly. Multidimensional scaling is adopted to allow converting multiple axes into a single subplot. The transition between two visual types is designed in a seamless way. In our work, a series of interaction tools has been developed. Uniform brushing functionality is implemented to allow the user to perform data selection on both points and parallel coordinate polylines without explicitly switching tools. A GPU accelerated Dimensional Incremental Multidimensional Scaling (DIMDS) has been developed to significantly improve the system performance. Our case study shows that our scheme is more efficient than traditional multi-view methods in performing visual analysis tasks.

show abstract

“…Visualizing a larger data set will inevitably result in a cluttered image which impairs the analysis process. Reducing clutter in parallel coordinates has, therefore, attracted the attention of many researchers and several interesting approaches have been proposed [11,12,13,14,15].…”

Section: Background and Related Workmentioning

confidence: 99%

“…The lines were rendered using transparency in order to make overlaps more visible. This is a common technique which is extensively used today and was first explored together with parallel coordinates by Wegman and Luo [12]. Depending on how two axes were positioned with respect to each other, a stimulus pattern could be mirrored.…”

Section: Patternsmentioning

confidence: 99%

Perceiving Patterns in Parallel Coordinates: Determining Thresholds for Identification of Relationships

Johansson

Forsell

Lind

et al. 2008

Information Visualization

View full text Add to dashboard Cite

This article presents a study that investigates the ability of humans to perceive relationships (patterns) in parallel coordinates, an ability that is crucial to the use of this popular visualization technique. It introduces a visual quality metric, acceptable distortions of patterns, which establishes the level of noise that may be present in data while allowing accurate identification of patterns. This metric was used to assess perceptual performance of standard 2D parallel coordinates and multi-relational 3D parallel coordinates in two experiments. In multi-relational 3D parallel coordinates the axes are placed on a circle with a focus axis in the centre, allowing a simultaneous analysis between the focus variable and all other variables. The experiments aimed to determine the maximum number of variables that can be, from a user's point of view, efficiently used in a multi-relational 3D parallel coordinates display and to present a first attempt to study users' ability to analyse noisy data in parallel coordinates. The results show that, in terms of the acceptable level of noise in data, a multi-relational 3D parallel coordinates visualization having 11 axes (variables) is as efficient as standard 2D parallel coordinates. Visualizing a larger number of variables would possibly require a greater amount of manipulation of the visualization and thus be less efficient.

show abstract

High Dimensional Clustering Using Parallel Coordinates and the Grand Tour

Cited by 82 publications

References 4 publications

Probability state modeling theory

Probability state modeling theory

Scattering Points in Parallel Coordinates

Perceiving Patterns in Parallel Coordinates: Determining Thresholds for Identification of Relationships

Contact Info

Product

Resources

About