Selecting Coherent and Relevant Plots in Large Scatterplot Matrices

Lehmann, Dirk J.; Albuquerque, Georgia; Eisemann, Martin; Magnor, Marcus; Theisel, Holger

doi:10.1111/j.1467-8659.2012.03069.x

Cited by 34 publications

(25 citation statements)

References 36 publications

(44 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the benefit of our approach is the enormous compression we achieve by collapsing similarity searches from O(pn) to O(p) through the use of scagnostics. Moreover, we implemented a quick clustering algorithm [14] with the complexity O(p) or O(v 2 ) compared to the greedy variable reordering algorithm [19] which requires O(v 3 ) (where v is the number of variables in the data). This provides ScagExplorer with the scalability to handle huge datasets up to thousands of dimensions.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

ScagExplorer: Exploring Scatterplots by Their Scagnostics

Dang

Wilkinson

2014

2014 IEEE Pacific Visualization Symposium

View full text Add to dashboard Cite

A scatterplot displays a relation between a pair of variables. Given a set of v variables, there are v(v − 1)/2 pairs of variables, and thus the same number of possible pairwise scatterplots. Therefore for even small sets of variables, the number of scatterplots can be large. Scatterplot matrices (SPLOMs) can easily run out of pixels when presenting high-dimensional data. We introduce a theoretical method and a testbed for assessing whether our method can be used to guide interactive exploration of high-dimensional data. The method is based on nine characterizations of the 2D distributions of orthogonal pairwise projections on a set of points in multidimensional Euclidean space. Working directly with these characterizations, we can locate anomalies for further analysis or search for similar distributions in a "large" SPLOM with more than a hundred dimensions. Our testbed, ScagExplorer, is developed in order to evaluate the feasibility of handling huge collections of scatterplots.

show abstract

Section: Discussionmentioning

confidence: 99%

“…In other words, it does not guarantee that the two adjacent plots are similar. Lehmann et al [19] used a heuristic optimization algorithm to reorder dimensions based on a scagnostics measure. This reordering method concentrates on the best plots in different regions, based on the similarity of the dimensions.…”

Section: Clustering Algorithmmentioning

confidence: 99%

ScagExplorer: Exploring Scatterplots by Their Scagnostics

Dang

Wilkinson

2014

2014 IEEE Pacific Visualization Symposium

View full text Add to dashboard Cite

show abstract

“…Many scagnostics techniques also use the small-multiple metaphor to show interesting projections, where interest is defined in a wide variety of ways [48][49][50][51]. Besides using the exploration history, the views V(P i ) can be picked by optimizing for diversity of this set, by subsampling an initial large random sampling of V, or by iteratively refining a given set of views to improve their diversity [52].…”

Section: Navigating Multidimensional Data Visualizationsmentioning

confidence: 99%

Multidimensional Data Exploration by Explicitly Controlled Animation

et al. 2017

View full text Add to dashboard Cite

Abstract:Understanding large multidimensional datasets is one of the most challenging problems in visual data exploration. One key challenge that increases the size of the exploration space is the number of views that one can generate from a single dataset, based on the use of multiple parameter values and exploration paths. Often, no such single view contains all needed insights. The question thus arises of how we can efficiently combine insights from multiple views of a dataset. We propose a set of techniques that considerably reduce the exploration effort for such situations, based on the explicit depiction of the view space, using a small multiple metaphor. We leverage this view space by offering interactive techniques that enable users to explicitly create, visualize, and follow their exploration path. This way, partial insights obtained from each view can be efficiently and effectively combined. We demonstrate our approach by applications using real-world datasets from air traffic control, software maintenance, and machine learning.

show abstract

“…They also present a scheme that allows to locally analyze the quality of distinct projection methods as well as the quality of clusters. Lehmann et al [18] present an interactive approach for generating ranked scatter plot matrices, where the elements are organized according to a reordering technique supported by quality measures. Lehmann's approach bears a set properties that are not present simultaneously in any other scatter plot-based method.…”

Section: Related Workmentioning

confidence: 99%