Data mining with its application in oil and gas exploration and exploitation /

Files

1287622.pdf (10.63 MB)
  (Published version)

Date

2009

Authors

Hao, Yalei,

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

thesis

Citation

Statement of Responsibility

Conference Name

Abstract

The petroleum industry has been involved in the application of artificial intelligence (AI) techniques since their inception. The recent research progress in Data Mining illuminates the combination of data mining and petroleum data analysis. In this work, we develop two novel classification algorithms. One is MOUCLAS (MOUntain function based CLASsification) mining, which integrates clustering and association mining to derive implicit relationships between condition and decision attributes. The other is an ensemble of multiple attribute selection measures for decision tree building, with the advantages of smaller tree sizes and greater accuracy than the original algorithms. The aim of these studies is to use novel classification algorithms to interpret the pay zones from well logging data for the purpose of reservoir characterization. We also develop a framework of knowledge management, which automatically deals with the utilization and involving of the prior and domain knowledge. These techniques have important implications, not only for intelligent well logging analysis, but also for many other applications.

School/Discipline

University of South Australia. School of Computer and Information Science.
School of Computer and Information Science.

Dissertation Note

Thesis (PhD(Information Technology))--University of South Australia, 2009.

Provenance

Copyright 2009 Yalei Hao.

Description

1 ethesis (vi, 110 pages) :
illustrations (some colour), charts (some colour)
Includes bibliographical references (p. 101-110)

Access Status

506 0#$fstar $2Unrestricted online access

Rights

License

Grant ID

Published Version

Call number

Persistent link to this record