OVERVIEW

Data Mining is usually used to describe the process of using advanced statistical techniques and/or artificial intelligence to discover knowledge from very large data sets, often with the aid of specialized tools. Data Mining tools are very powerful. But as with all powerful tools they can be dangerous in the wrong hands, careful preparation is essential and sometimes it is safer to use a less powerful tool. In this seminar we will look at a number of different applications ranging from the traditional to new areas including e-commerce and multimedia. A commonly cited example is that a supermarket chain using knowledge discovery found an unexpected link between the sales of beer and nappies during the early evening period. Having determined the link they could then experiment with positioning the good so as to maximize sales. E-commerce companies are contemplating different pricing policies depending on the history of the customer. Banks use predictive data mining, they use past data to determine the characteristics of bad credit risks and then apply this discovered knowledge to decide whether or not to lend money to new clients. Multimedia systems can be designed so that knowledge of past usage of the system can be used to predict future access and so reduce the wait time for downloads.

AUDIENCE

This seminar is intended to provide professional software engineers with an insight into the knowledge discovery process and the tools available for data mining. It will be of particular interest to those who have not previously been involved in a data mining project but may in the future be called upon to scope such a project.

ABOUT THE PRESENTER

Shirley Williams, B.Sc, PhD, FBCS, CEng, is a senior lecturer at the University of Reading, teaching in the area of Software Engineering. She spent the academic year 1999-2000 on sabbatical leave working with Vodafone working on two projects: Building a knowledge base for dimensioning exchanges Knowledge discovery with engineering applications.

VISITOR COUNT