Skip to main content

Data mining and database systems: integrating conceptual clustering with a relational database management system.

Lepinioti, K.., 2011. Data mining and database systems: integrating conceptual clustering with a relational database management system. Doctoral Thesis (Doctoral). Bournemouth University.

Full text available as:

[img]
Preview
PDF
Lepinioti,Konstntina_Ph.D.-2011_-embargo_until_May_2012.pdf

2MB

Abstract

Many clustering algorithms have been developed and improved over the years to cater for large scale data clustering. However, much of this work has been in developing numeric based algorithms that use efficient summarisations to scale to large data sets. There is a growing need for scalable categorical clustering algorithms as, although numeric based algorithms can be adapted to categorical data, they do not always produce good results. This thesis presents a categorical conceptual clustering algorithm that can scale to large data sets using appropriate data summarisations. Data mining is distinguished from machine learning by the use of larger data sets that are often stored in database management systems (DBMSs). Many clustering algorithms require data to be extracted from the DBMS and reformatted for input to the algorithm. This thesis presents an approach that integrates conceptual clustering with a DBMS. The presented approach makes the algorithm main memory independent and supports on-line data mining.

Item Type:Thesis (Doctoral)
Additional Information:If you feel that this work infringes your copyright please contact the BURO Manager.
Group:UNSPECIFIED
ID Code:17765
Deposited By:INVALID USER
Deposited On:11 May 2011 15:19
Last Modified:09 Aug 2022 16:03

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -