Visual Neural Decomposition to Explain Multivariate Data Sets

  title={Visual Neural Decomposition to Explain Multivariate Data Sets},
  author={Johannes Knittel and Andr{\'e}s Lalama and Steffen Koch and Thomas Ertl},
  journal={IEEE Transactions on Visualization and Computer Graphics},
Investigating relationships between variables in multi-dimensional data sets is a common task for data analysts and engineers. More specifically, it is often valuable to understand which ranges of which input variables lead to particular values of a given target variable. Unfortunately, with an increasing number of independent variables, this process may become cumbersome and time-consuming due to the many possible combinations that have to be explored. In this paper, we propose a novel… 

Multivariate Data Explanation by Jumping Emerging Patterns Visualization

This paper presents multiVariate dAta eXplanation (VAX), a new VA method to support identifying and visual interpreting patterns in multivariate datasets that uses the concept of Jumping Emerging Patterns, inherent interpretable logic statements representing class-variable relationships ( patterns) derived from random Decision Trees.

A Visual Analytics System for Water Distribution System Optimization

A visual analytics system that helps understand relationships between simulation inputs and outputs towards better optimization and can lessen the burden of analysis and assist in determining optimal operating schemes is designed.

Intelligent Methods for Test and Reliability

The overall philosophy of the Graduate School on “Intelligent Methods for Semiconductor Test and Reliability” and the specific scientific questions targeted by its ten projects are described.



Explainers: Expert Explorations with Crafted Projections

  • Michael Gleicher
  • Computer Science
    IEEE Transactions on Visualization and Computer Graphics
  • 2013
An approach to exploration and discovery in high-dimensional data that incorporates a user's knowledge and questions to craft sets of projection functions meaningful to them and considers tradeoffs in choosing these projection functions, including their simplicity, expressive power, alignment with prior knowledge, and diversity.

Subspace search and visualization to make sense of alternative clusterings in high-dimensional data

  • A. TatuFabian Maass D. Keim
  • Computer Science
    2012 IEEE Conference on Visual Analytics Science and Technology (VAST)
  • 2012
A novel method is proposed for the visual analysis of HD data in which it is proposed that an interestingness-guided subspace search algorithm is employed to detect a candidate set of subspaces and, based on appropriately defined subspace similarity functions, this approach allows users to effectively compare and relate sub Spaces with respect to involved dimensions and clusters of objects.

Visual‐interactive Exploration of Interesting Multivariate Relations in Mixed Research Data Sets

A visual‐interactive system that supports scientists to explore interesting relations between aggregated bins of multivariate attributes in mixed data sets and enables the application of statistical dependency tests as the measure of interestingness is presented.

Supporting Analysis of Dimensionality Reduction Results with Contrastive Learning

This work presents a visual analytics method that effectively highlights the essential features of a cluster in a DR result and introduces an enhanced usage of contrastive principal component analysis (cPCA), called ccPCA (contrasting clusters in PCA).

LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in Recurrent Neural Networks

This work presents LSTMVis, a visual analysis tool for recurrent neural networks with a focus on understanding these hidden state dynamics, and describes the domain, the different stakeholders, and their goals and tasks.

INFUSE: Interactive Feature Selection for Predictive Modeling of High Dimensional Data

INFUSE is a novel visual analytics system designed to help analysts understand how predictive features are being ranked across feature selection algorithms, cross-validation folds, and classifiers, and it is demonstrated how the system can lead to important insights in a case study involving clinical researchers predicting patient outcomes from electronic medical records.

SeekAView: An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces

The proposed SeekAView integrates a visually scalable design space with an iterative workflow guiding the analysts by choosing the starting points and letting them slice and dice through the data to find interesting subspaces and detect correlations, clusters, and outliers.

Visual Correlation Analysis of Numerical and Categorical Data on the Correlation Map

This work devise mechanisms that handle both categorical and numerical variables within a unified framework that achieve scalability for large numbers of variables via a multi-scale semantic zooming approach and provides interactive techniques for exploring the impact of value bracketing on correlations.

Representative Factor Generation for the Interactive Visual Analysis of High-Dimensional Data

This paper introduces the construction and utilization of representative factors for the interactive visual analysis of structures in high-dimensional datasets, and presents a selection of methods to investigate the sub-groups in the dimension set and associate representative factors with those groups of dimensions.

Visualizing Data using t-SNE

A new technique called t-SNE that visualizes high-dimensional data by giving each datapoint a location in a two or three-dimensional map, a variation of Stochastic Neighbor Embedding that is much easier to optimize, and produces significantly better visualizations by reducing the tendency to crowd points together in the center of the map.