WIT Press


Data Warehouse Screening And Personalizing Agent

Price

Free (open access)

Paper DOI

10.2495/DATA040031

Volume

33

Pages

9

Published

2004

Size

240 kb

Author(s)

A. H. Mohamed, N. L. M. Noor & N. M. Noh

Abstract

Intelligent software agents have been developed for many purposes these days. Software agents have been developed in many types of applications such as Internet search engines etc. This article is all about implementing the intelligent software agent in a data warehouse implementation. The first purpose of this software agent is to screen data in order to avoid data duplication in the data warehouse. The second objective of this research is to personalize the user requirements of data marts in data warehouse environment. The software agent consists of two basic modules. The first module is the information screening module. This module is implemented as an addition to the data staging process, before the data is loaded into the data warehouse. It functions as a pre-loading step, where the data to be loaded into the data warehouse has been cleaned, extracted and transformed. This module uses an enhanced Fuzzy String Matching to check the existence of data in the data warehouse. The second module involved is the personalization module. This is a module where the agent will study the history queries sent by users to the system. The queries may represent user requirements to find any pattern of time of the queries being sent to the system. The technique used by the agent to study users’ queries is grouping the queries which are similar to each other. The largest group of queries is considered as the most needed query. However, this agent may be enhanced to produce a more robust intelligent agent. Keywords: intelligent software agent, data warehouse, information screening, personalization.

Keywords

intelligent software agent, data warehouse, information screening, personalization.