Applied Information Management

Visualizing Big Data

The sheer volume of data and information that require handling, processing, and analysis is growing at an exponential rate across industries. The scale and scope of changes that big data brings are greatly expanding as technology accelerates structured and unstructured data collection; these changes provide increasing information accessibility and management challenges when attempting to derive meaning, enable sense-making, and generate knowledge. Decision support systems are typically used to support the communication of inherent meaning within large data sets. The overall goal of information visualization, regardless of implementation environment, is to enable rapid cognition without the manual scanning of detailed data sets often found in traditional information presentation solutions such as decision support systems.

Information visualizations set the context for data exploration.

This study identifies (a) information visualization techniques that can support the exploration, discovery, and crystallization of new knowledge based on visual recognition of hidden patterns and structures in big data, and (b) evaluation criteria that can be used to determine the effectiveness of information visualization within a specific information domain. References presented in an annotated bibliography focus on one of four aspects of information visualization: (a) interaction, navigation, and exploration; (b) discovery and practices; (c) design and implementation; and (d) decision support and effectiveness evaluation. A concept map (see Figure 1) demonstrates the relationships among these four aspects.

Color-coded information visualization concept map node groupings
Figure 1. Color-coded information visualization concept map node groupings

The primary audience for this annotated bibliography is software development teams challenged with designing and implementing information visualization software that handles large data sets. Figure 2 (from Keim, 2000) provides considerations regarding the categories of visualization techniques best suited for the overall complexity of the data set(s) under investigation.

Chart of Dimensions of Exploratory Data Visualizations
Figure 2. Dimensions of exploratory data visualizations

Data exploration is defined as the process of searching and analyzing databases to find implicit but potentially useful information. Keim (2002) presents a useful tutorial on explorative analysis, noting that such analysis consists of (a) establishing a starting point, (b) defining a process, and (c) generating a result; he provides examples of visualization techniques for each classification, including circle segments and cone trees (see Figures 3 and 4).

Example circle segments. A circle segments chart demonstrating visualization pixel-oriented techniques. This chart is an example only.
Figure 3. Circle segments. This visualization demonstrates pixel-oriented techniques.
Example cone tree. A cone tree chart demonstrating visualization hierarchical techniques. This chart is an example only.
Figure 4. Cone trees. This visualzation demonstrates hierarchical techniques.

Selected References

  • Fisher, D., DeLine, R., Czerwinski, M., & Drucker, S. (2012). Interactions with big data analytics. Retrieved from ACM Digital Library.
  • Keim, D. (2000). An introduction to information visualization techniques for exploring large databases. Retrieved from Swiss Federal Institute of Technology Zurich.
  • Li, T., Feng, S., & Li, L. (2001). Information visualization for intelligent decision support systems. Knowledge-Based Systems, 14(5-6), 259-262.
  • Manyika, J., Chui, M., Brown, B., Bughin, J., Dobbs, R., Roxburgh, C., & Byers, A. (2011). Big data: The next frontier for innovation, competition, and productivity. Retrieved from McKinsey and Company.
  • Mouine, M., & Lapalme, G. (2012). Using clustering to personalize visualization (Ed). 2012 16th International Conference on Information Visualization (pp. 258). nd: IEEE. doi: 10.1109/IV.2012.51.
  • Offenhuber, D. (2010). Visual anecdote. Leonardo, 43(4), 367-374.
AIM alumnus Brian Fabrizio

Research Paper Author: Brian Fabrizio, knowledge base analyst, Givaudan—2013 University of Oregon, AIM Program Graduate.

Abstract: This annotated bibliography examines information visualization concepts and techniques supporting exploration, discovery, and crystallization of new knowledge based on visual recognition of hidden patterns and structures in big data (Offenhuber, 2010), and effectiveness evaluation methodologies within specific domains. Literature published between 2000–2012 suggests considerations for software development teams tasked with creating information visualization solutions. Results include a taxonomy of terms and concepts, a concept map, and over a dozen images demonstrating visualization techniques.