In this paper, we provide an overview of several pixel-oriented visualization techniques which have been developed over the last years to support an effective querying and exploration of large databases. Pixel-oriented techniques use each pixel of the display to visualize one data value and therefore allow the visualization of the largest amount of data possible. The techniques may be divided into query-independent techniques which directly visualize the data (or a certain portion of it) and query-dependent techniques which visualize the relevance of the data with respect to a specific query. An example for the class of query-independent techniques is the recursive pattern technique which is based on a generic recursive scheme generalizing a wide range of pixel-oriented arrangements for visualizing large databases. Examples for the class of query-dependent techniques are the generalized spiral and circle-segments techniques, which visualize the distances with respect to a database quer...

An important goal of visualization technology is to support the exploration and analysis of very large amounts of data. In this paper, we describe a set of pixeloriented visualization techniques which use each pixel of the display to visualize one data value and therefore allow the visualization of the largest amount of data possible. Most of the techniques have been specifically designed for visualizing and querying large databases. The techniques may be divided into query-independent techniques which directly visualize the data (or a certain portion of it) and query-dependent techniques which visualize the data in the context of a specific query. Examples for the class of query-independent techniques are the screen-filling curve and recursive pattern techniques. The screen-filling curve techniques are based on the well-known Morton and Peano-Hilbert curve algorithms, and the recursive pattern technique is based on a generic recursive scheme which generalizes a wide range of pixel-ori...

Daniel A. Keim Institute for Computer Science, University of Munich Leopoldstr. 11 B, D-80802 Munich, Germany, Phone (+49) 89 2180-6267 keim@informatik.uni-muenchen.de 1. Introduction An important goal of visualization technology is to support the exploration and analysis of very large amounts of data which are usually stored in databases. Since number and size of the databases is growing rapidly, there is a need for novel visualization techniques which allow a visualization of larger amounts of data. Most of today's databases store typical transaction-generated multiattribute data which does not have any inherent two- or three-dimensional semantics and therefore does not lend itself to some two- or three-dimensional visualization. In general, databases can be seen as multidimensional data sets with the attributes of the database corresponding to the dimensions of the multidimensional data set. There are a variety of well known techniques for visualizing arbitrary multidimensional da...

Visualizing hierarchical data is one of the core areas of information visualization. Most of these techniques focus on single hierarchies—hierarchies with a single root element and a single path to each element. In contrast, this work focuses on the browsing of multi-hierarchies—hierarchies with multiple roots or multiple paths per element. A radial focus+context display algorithm and interaction methods are introduced to explore such multi-hierarchical data. A series of examples demonstrate the effectiveness of our new visualization.

This paper describes an interactive data exploration system for molecular and clinical data in the field of personalized medicine. It addresses the essential but to date unsolved problem of how to identify connections between genetic variants and their corresponding diseases or the response to certain drugs and treatments, respectively. It is therefore necessary to connect genetic with clinical data in order to categorize specific subgroups of patients with certain disease features. The huge amount of data provided by molecular analytical methods (e.g. data on genetic alterations, proteomic or metabolomic data) can only be analyzed by applying statistical methods and bioinformatics. However, even standard methods of statistics and bioinformatics fail when the data is inhomogeneous – as is the case with clinical data – and when data structures are obscured by noise and dominant patterns. The structure of large medical data sets is made visible by using so called object- and attribute-glyphs, which can be arranged in a two dimensional space and synchronized with a set of visualization views. 1.

We present a visual data mining application using the combination of clinical data, pathways and biomolecular data. Using pathways to navigate and filter the clinical and molecular data allows a more systematic and efficient investigation of problems in modern life science. A multiplicity of hypotheses can be evaluated in the same period of time, enabling a much better exploitation of the data. We present a system for data preprocessing and automatic classification, a set of visualization views and finally the integration of the views in the Caleydo visualization framework, which enables the “coupling ” of molecular and a broad spectrum of clinical data. With the help of the Caleydo framework the medical expert can identify connections between genetic parameters, patient subgroups, and drug responses in an intuitive way.

Pixel-oriented visualization techniques map each attribute value of the data to a single colored pixel, yielding the display of the most possible information at a time. Thus pixel-oriented techniques maintain the global view of large amounts of data while still preserving the perception of small regions of interest. This property makes them suitable for a variety of data mining tasks. First we present pixel-oriented visualization techniques which can be used as stand-alone exploration tools. Then we show how they can be tightly integrated into data mining methods unifying the strength of existing algorithms and human involvement. Finally, we point out the idea of similarity clustering of attributes to enhance multidimensional visualization techniques.

Abstract not found