(PHBS)

Publications

Home > Faculty & Research > Publications >

Blockmap: An Interactive Visualization Tool for Big-Data

by Terrill L. FRANTZ

ARTICLE | Computational and Mathematical Organiation Theory | Vol.24, 2018


Abstract


This article describes the Blockmap, which is a mechanism for displaying and exploring network datasets. The data are presented in a squarified-mosaic form, which is well-suited for visual display on a computer or phone screen. The relational data are dimension-reduced and structured for interactive, hierarchical exploration. The Blockmap applies a combination of treemap and heatmap display schemes specifically to the analysis of large network datasets. The Blockmap offers the analyst a way to explore underlying node-level data, at the full-network level, according to shared characteristics of the constituent nodes. It offers a technique for exploring nodesets—collections of network nodes—which have been classified according to a user-defined set of rules or discriminative algorithms. Typically, nodes can be classified according to their common attributes or a stratification of their ego-level network measures, but means can be extended. Using a Blockmap, an analyst can profile a network according to the meaningful characteristics exhibited by the mosaic; this technique also offers theorists a platform for developing a methodological and analytic framework for characterizing and analyzing network data. Production versions of Blockmap technology are presently hosted in client- and web-based software and is available freely in *ORA-LITE.
Popular Articles
Latest News
Campus Events