Computational Methods for High-Dimensional Rotations in Data Visualization

Buja, Andreas; Cook, Dianne; Asimov, Daniel; Hurley, Catherine B.

doi:10.1016/s0169-7161(04)24014-7

Cited by 53 publications

(60 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other exploration techniques help the user by giving controls for the projection matrix (e.g., Refs. [27,28]). Nam and Mueller [29] proposed a projection navigation interface, where the exploration and navigation activities are decomposed into five major tasks: sight identification, tour path planning, touring, looking around & zooming into detail, and orientation & localization.…”

Section: Visual Exploration Of High-dimensional Datamentioning

confidence: 99%

“…Thus, the layout of the glyphs is determined using a local affine multi-dimensional projection algorithm [42]. The user can plan a rotation path containing a sequence of projection glyphs among the multiple projection glyphs, then a rotation animation is generated based on interpolation between adjacent projections along the path [27] with a slider to control the position of the animation.…”

Section: Visual Exploration Of the Projected Scatterplotmentioning

confidence: 99%

See 1 more Smart Citation

EasySVM: A visual analysis approach for open-box support vector machines

Chen

et al. 2017

Comp. Visual Media

View full text Add to dashboard Cite

Support vector machines (SVMs) are supervised learning models traditionally employed for classification and regression analysis. In classification analysis, a set of training data is chosen, and each instance in the training data is assigned a categorical class. An SVM then constructs a model based on a separating plane that maximizes the margin between different classes. Despite being one of the most popular classification models because of its strong performance empirically, understanding the knowledge captured in an SVM remains difficult. SVMs are typically applied in a black-box manner where the details of parameter tuning, training, and even the final constructed model are hidden from the users. This is natural since these details are often complex and difficult to understand without proper visualization tools. However, such an approach often brings about various problems including trial-and-error tuning and suspicious users who are forced to trust these models blindly.The contribution of this paper is a visual analysis approach for building SVMs in an open-box manner. Our goal is to improve an analyst's understanding of the SVM modeling process through a suite of visualization techniques that allow users to have full interactive visual control over the entire SVM training process. Our visual exploration tools have been developed to enable intuitive parameter tuning, training data 2016-12-24; accepted: 2017-01-09 manipulation, and rule extraction as part of the SVM training process. To demonstrate the efficacy of our approach, we conduct a case study using a real-world robot control dataset.

show abstract

Section: Visual Exploration Of High-dimensional Datamentioning

confidence: 99%

Section: Visual Exploration Of the Projected Scatterplotmentioning

confidence: 99%

EasySVM: A visual analysis approach for open-box support vector machines

Chen

et al. 2017

Comp. Visual Media

View full text Add to dashboard Cite

show abstract

“…A number of different algorithms to implement the Grand Tour have been developed over time and their description is beyond the scope of this paper. The reader is referred to (Wegman et al, 2002), where a discussion of several approaches can be found, and to a more recent publication (Buja et al, 2005) on Grand Tours and related data visualisation methods.…”

Section: The Grand Tourmentioning

confidence: 99%

“…A framework to generate a comprehensive range of low-dimensional projections is the Grand Tour proposed by Diane Cook and colleagues (Buja et al, 2005;Cook and Swayne, 2007). In this approach, the data are effectively tumbled in a systematic way and viewed through the prism of low-dimensional linear projections, looking for indicators of structure, typically gaps between sub-population cohorts.…”

Section: Introductionmentioning

confidence: 99%

“…Due to the usefulness of the visualisations obtained, and the need of an easy and straightforward way to obtain them, this method has been recently implemented in an R package (Wickham et al, 2011), with a user-friendly graphical user interface (GUI) (Huang et al, 2012). While this approach is powerful in principle, we reckon that the search procedure may be expedited by prioritising the most informative views of the data; this is the goal and main novelty of the proposed approach compared to (Buja et al, 2005;Cook and Swayne, 2007). In (Lecerf and Bouchard, 2012), a method based on selecting candidate projections from the space of all projections was proposed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Making nonlinear manifold learning models interpretable: The manifold grand tour

Lisboa

Martín-Guerrero

Vellido

2015

Expert Systems with Applications

View full text Add to dashboard Cite

Dimensionality reduction is required to produce visualizations of high dimensional data. In this framework, one of the most straightforward approaches to visualising high dimensional data is based on reducing complexity and applying linear projections while tumbling the projection axes in a defined sequence which generates a Grand Tour of the data. We propose using smooth nonlinear topographic maps of the data distribution to guide the Grand Tour, increasing the effectiveness of this approach by prioritising the linear views of the data that are most consistent with global data structure in these maps. A further consequence of this approach is to enable direct visualisation of the topographic map onto projective spaces that discern structure in the data. The experimental results on standard databases reported in this paper, using Self-Organising Maps and Generative Topographic Mapping, illustrate the practical value of the proposed approach. The main novelty of our proposal is the definition of a systematic way to guide the search of data views in the grand tour, selecting and prioritizing some of them, based on nonlinear manifold models.

show abstract

References

2012

Methods of Multivariate Analysis

View full text Add to dashboard Cite

Computational Methods for High-Dimensional Rotations in Data Visualization

Cited by 53 publications

References 8 publications

EasySVM: A visual analysis approach for open-box support vector machines

EasySVM: A visual analysis approach for open-box support vector machines

Making nonlinear manifold learning models interpretable: The manifold grand tour

References

Contact Info

Product

Resources

About