Project Description:
Breakdown Visualization is a computer based visualization system for financial
analysts. The tool provides support for browsing, querying, benchmarking,
and pattern cause identification from financial statement databases such
as COMPUSTAT. Managers and financial analysts continue to utilize
financial statements, reports, and financial ratios to help them in their
daily decision making. DuPont Analysis is a tool available to help
these decision makers to determine cause-effect relationships and to find
key indicators. Breakdown Visualization is a tool that supports this
analysis, helping managers better identify business problems and successes.
"The most important thing to remember with respect to these issues is that
ratio analysis doesn't give answers; it helps you ask the right questions"
(Lasher, William R. Practical Financial Management, Southwestern
Publishing, 2000. p. 87).
Breakdown analysis involves decomposing data into sub-groups to allow
for comparison and identification of problem areas. Good analysis
requires the ability to group data based on attributes or values.
Breakdown Visualization provides a mechanism to support this analysis through
user guided decomposition and exploration of tabular data with a polyarchical
structure. This is useful in many domains, including sports statistics
and corporate financial reports. Breakdown Visualization utilizes
a visual spreadsheet format for comparison of adjacent visualizations.
The work builds on existing ideas introduced in the Polyarchy
Visualization project at Microsoft Research (George
Robertson et. al.).
Participants:
Chris North, Assistant Professor
- Computer Science
Nathan Conklin, Graduate
Student - Computer Science and Business Administration
Sandeep Prabhakar, Graduate Student
- Computer Science
Muthukumar Thirunavukkarasu, Graduate
Student - Computer Science
Papers and Presentations:
-
Conklin, N., Prabhakar, S., and North, C., "Multiple
Foci Drill-Down trough Tuple and Attribute Aggregation Polyarchies in Tabular
Data", Proceedings of IEEE InfoVis 2002 Symposium, (October
2002).
-
Conklin, N., Prabhakar, S., and North, C., "Multiple
Foci Drill-Down trough Tuple and Attribute Polyarchies in Tabular Data",
TR-02-12, Computer Science Department, Virginia Polytechnic Institute
and State University, (June 2002).
-
Conklin, N., Prabhakar, S., and North, C., "Breakdown
Visualization: Multiple Foci Polyarchies of Tuples and Attributes",
Presentation, Proceedings of ACM CHI, March 2002.
-
Prabhakar, S., Conklin, N., North, C., Thirunavukkarasu, M., Dandapani,
A., and Panchanathan, G., "Breakdown
Visualization: Multiple Foci Polyarchies of Tuples and Attributes",
Proceedings
of ACM CHI, March 2002.
-
Conklin, N., "DuPont Visualization", Independent Study Report, Computer
Science Department, Virginia Polytechnic Institute and State University,
August 2001.
Future Work
Future work will focus on supporting the need to allow users to define
their own heirarchy, save and publish an analysis, support a feed-forward
affordance that assists analysis, and the notion of defining data primitives.
