home · mobile · calendar · defenses · 2007-2008 · 

Thesis Defense - Hashemi

The Analytical Mediator for Multi-Dimensional Data
Computer Science PhD Candidate

As corporations have been producing -- and keeping -- increasing volumes of electronic data, there has been a growing interest in analyzing legacy data in hopes of making future business decisions more profitable. Thus, data mining and statistical data analyzing has been a growing field, especially in the financial, health care and software industries. But, while numerous, sophisticated, generalized mathematical algorithms for analyzing data have been developed in recent years, many of the practical problems associated with long term data analysis have remain largely unaddressed.

The Analytical Mediator System (AMS) is an experimental prototype system aimed directly at this critical research need. In particular, it addresses the specific processes of importing, cleansing, transforming and integrating, and more importantly, for performing practical analysis of heterogeneous data. The research behind AMS was based on creating a flexible tool coupled with experimenting different data analysis techniques to produce more combinations from data sets.

The AMS provides a straightforward Hosting application model for managing the entire data management lifecycle. Other valuable features of the AMS include its facility for constructing a domain model, its ability to interface with some tools, its generation of comprehensible reports and its easy-to-use facility for reducing the dimensionality of a large data set.

Committee: Roger (Buzz) King, Professor (Chair)
Richard Byrd, Professor
Andrzej Ehrenfeucht, Distinguished Professor
Karen Kafadar, University of Colorado at Denver
Richard Osborne, (Former) Assistant Professor
Department of Computer Science
University of Colorado Boulder
Boulder, CO 80309-0430 USA
May 5, 2012 (14:20)