University of Houston

Department of Computer Science

In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy

Wei Ding

Will defend her thesis proposal

Mining Regional Knowledge in Spatial Datasets

Abstract

The immense explosion of geographically referenced data calls for efficient discovery of spatial knowledge. A special challenge for spatial data mining is that information is usually not uniformly distributed in spatial datasets. Consequently, the discovery of regional knowledge is of fundamental importance for spatial data mining. Unfortunately, traditional data mining techniques are ill-prepared for discovering regional knowledge. This raises the questions on how to measure the interestingness of a set of regions and how to search effectively and efficiently for interesting regional knowledge. This proposal proposes a framework to systematically discover regional knowledge in spatial datasets. Different measures of interestingness are introduced. Grid-based and density-based algorithms will be implemented to identify interesting regions.

In our preliminary research, we have implemented a methodology for mining regional association rules and developed new algorithms to determine the scope of a regional association rule. We solved two distinct region discovery problems: identifying interesting regions for regional association rule mining, and determining the scope where a regional association rule holds. Our association rule mining methodology is supervised, centering on finding rules with respect to an underlying class structure. The class structure itself is also used for rule pruning and for the discretization of numerical attributes. We will present our experimental results in a real-world case study to identify spatial risk patterns from arsenic in the Texas water supply.

Date: Wednesday, November 15, 2006
Time: 10:00 AM
Place: 550 PGH

Faculty, students, and the general public are invited.
Advisor: Dr. Christoph F. Eick