Data mining applications are being utilized in a number of fields. A
few specific applications from different areas follows next:
Marketing:
A number of marketers are in the opinion that one of the most
important means within the business is to be able to understand each
customers individual needs. Data mining technology is being used to
shed a light on the customers preferences and buying patterns. The
program Spotlight from A.C. Nielsen in Illinois can be used for
this. Spotlight can search databases of Terra-byte-size with sales data
for thousands of products, scattered over hundreds of geographical
areas over long periods of time. A job which would have taken a human
weeks or months to do, gets done by Spotlight in minutes or hours. The
results can be used to determine which persons to offer products
and sales information in order to maximize profit.
Health:
GTE Laboratories has made Health-KEFIR, an advanced data mining
system which finds areas where costs are likely to increase in
hospitals, and among these select where specific actions probably
will save most money. Instead of a thick report, KEFIR gives the
results as HTML- and GIF-files, easily accessible through WEB-client
software.
Science:
Data mining techniques have started to assist humans in doing
scientific discoveries. By traversing enormous datasets, patterns are
found in molecular structures, genetic data, global climatic changes,
and more. SKICAT (SKy Image Cataloging and Analysis Tool) is an
advanced data mining system which automatically analyzes and
catalogizes observations the Palomar telescope has done over the night
sky in the northern hemisphere. SKICAT recently discovered nine
quasars in a fraction of the observation time it otherwise would have
taken.
Finance:
The world of finance is naturally reluctant in giving out information
regarding applications which give competitive advantages. Still, there
has been reported a number of applications which uses data mining
techniques. Several investment companies, for instance, make stock
transactions on the basis of data mining systems. Others have
implemented systems aimed at detecting and preventing fraud.
The Artificial Intelligence Research group at Lockheed Martin has been investigating and developing data mining tools for the past 10 years. Recently, the group built an internal application-development tool, called Recon, that generalizes their data mining techniques, and then applied it to application-specific problems (in particular finance). The system has an open architecture, running on Unix platforms and massively parallel supercomputers. It interfaces with existing relational database management systems, financial databases, proprietary databases, data feeds, spreadsheets, and ASCII files.