Masters Theses

Shipra Dutta

Keywords and Phrases

Hybrid Clustering Architecture

Abstract

"Data mining is a very active research area, which has gained a lot of attention over the past few years for its applications in various fields ranging from stock forecasting to detecting damages in the structures like aircrafts. The use of data mining for applications in science and business has evoked a lot of interest in the researchers to explore the area further. In this research, a novel hybrid clustering architecture will be presented as a possible solution to classify the geo-chemical data obtained by conducting various direct and indirect tests on different sites. The fact that no single learning algorithm will do well for all domains has stimulated a great deal of research in the area of combining algorithms.

The geo-chemical data pertains to the presence, amount and distribution of surface hydrocarbons. The processing of this geo-chemical data is of vital importance for detecting the types and amount of pollutants present at a particular site, which further helps in formulating sound environmental remediation policies. The knowledge of the hydrocarbons can also be associated with the specific type of fuel present at a site, and this helps in formulating the policies for assessment of oil and gas resources. The traditional approaches for evaluation of the site's conditions for specific exploration and exploitation needs suffers from some drawbacks discussed later in this thesis. In addition, because of many complex interactions involved in the data attributes, the approaches of constructing hydrocarbon micro seepage models for processing the geo-chemical data is not well defined. So, the application of data mining to the geo-chemical data is studied, which may help to automate the process of extracting useful knowledge or complex associations in the given information"--Abstract p. iii

Advisor(s)

Rao, Vittal S.

Committee Member(s)

St. Clair, Daniel
Leopold, Jennifer

Department(s)

Computer Science

Degree Name

M.S. in Computer Science

Publisher

University of Missouri--Rolla

Publication Date

Fall 2003

Pagination

xii, 85 pages

Note about bibliography

Includes bibliographical references (pages 81-84)

Rights

© 2003 Shipra Dutta, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Subject Headings

Data mining
Cluster analysis -- Computer programs
Hydrocarbons -- Analysis

Thesis Number

T 8425

Print OCLC #

55215772

Download

Share

 
COinS