Royal Statistical Society
Statistical Computing Section Meetings
Quality Forum
Half-day joint meeting with the Quality Improvement Section
Steve Thornton (Corus Research, Development and Technology
)
"Pragmatic approach to data mining"
Synopsis: In the industrial context data mining plays an important role in the knowledge management process; not normally as a generator of new knowledge, but rather in the formalizing and aggregation of fragments of information and expert intuition. A data miner should be equally skilled in interviewing and listening as in data manipulation and analysis; thus the search for useful patterns in data needs to be an iterative process of collaboration. Discoveries more often emerge as clues about promising options rather than absolute solutions, but the end product should be validated knowledge that can be applied to achieve improved performance. Speed and productivity are key requirements for success in the data mining approach and supportive tools and techniques are essential. Illustrative examples from other industries, or even from the past also serve to engage business representatives in the challenge of 'making their data earn its keep'. This talk will expand upon these ideas and illustrate with examples from Corus.
David Hand (Imperial College, London
)
"Deception, distortion, and discovery: data quality
in data mining "
Synopsis: Data mining is typically a process of secondary data analysis, using data
which were originally collected for some other purpose. They may have
been of high quality for that purpose, but of low quality for the
unspecified future analyses of data mining, and it may be economically
impracticable to require high quality data for all possible future
analyses. This talk gives an overview of data quality, covering
definitions, measurement, monitoring, and improvement. Some important
special topics are discussed in detail, including missing values, anomaly
detection, and deliberate data distortion. The talk is illustrated with
real examples from a wide variety of areas.
Mark Robinson ( Marketing Data Basics, Edinburgh
)
"Applied data mining in financial services"
Synopsis: In financial services, predictive modeling and data mining are often used to direct marketing spend in an effort to increase customer value and persistency. This talk will demonstrate a framework around which customer contact can be optimized with the application of propensity modeling, segmentation and good common sense. Financial organizations are starting to appreciate the value of their customer data and making it work for them. Based on real life case studies Mark will cover how statistics, data mining and pragmatism improve customer satisfaction and profitability.
Date & Time
Thursday 23rd June 2005 at 2:00-5:00 pm
Place
Errol Street
Refreshments
There will be tea and biscuits served at about 3:45 pm.
