Visualization of directed associations in e-commerce transaction data

  • Authors:
  • Ming C. Hao;Umeshwar Dayal;Meichun Hsu;Thomas Sprenger;Markus H. Gross

  • Affiliations:
  • Hewlett Packard Research Laboratories, Palo Alto, CA;Hewlett Packard Research Laboratories, Palo Alto, CA;Hewlett Packard Research Laboratories, Palo Alto, CA;Department of Computer Science, Swiss Federal Institute of Technology, Zurich, Switzerland and Hewlett Packard Research Laboratories, Palo Alto, CA;Department of Computer Science, Swiss Federal Institute of Technology, Zurich, Switzerland and Hewlett Packard Research Laboratories, Palo Alto, CA

  • Venue:
  • EGVISSYM'01 Proceedings of the 3rd Joint Eurographics - IEEE TCVG conference on Visualization
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

Many real-world e-commerce applications require the mining of large volumes of transaction data to extract marketing and sales information. This paper describes the Directed Association Visualization (DAV) system that visually associates product affinities and relationships for large volumes of e-commerce transaction data. DAV maps transaction data items and their relationships to vertices, edges, and positions on a visual spherical surface. DAV encompasses several innovative techniques (1) items are positioned according to their associations to show the strength of their relationships; (2) edges with arrows are used to represent the implication directions; (3) a mass-spring engine is integrated into a visual data mining platform to provide a self-organized graph. We have applied this system successfully to market basket analysis and e-customer profiling Internet applications.