Overview of Data Mining - Java Data Mining: Strategy, Standard, and Practice - page 2

Chapter
1
Overview of Data Mining
The most incomprehensible thing about the world
is that it is at all comprehensible.
—Albert Einstein
Imagine one's surprise at discovering which genes determine
susceptibility to a certain type of cancer by running an algorithm on
data consisting of 5,000 genes from each of a hundred patients.
Imagine one's surprise at being able to predict with high accuracy
which customers will purchase a specific product. Einstein's com-
ment on comprehensibility fits well with the world of data mining.
What is so amazing is that by amassing data from the real world on
just about anything, patterns can be determined that provide
insights into the world the data represents, making the world more
comprehensible . Using data mining to gain insight into seemingly
random data points is an increasingly common strategy among
business analysts, scientists, and researchers.
Although the complexity of some data mining algorithms is great,
using them has been greatly simplified through automation and
higher level abstractions, such as those found in Java Data Mining 1.1
(JSR-73) [JSR-73 2004]. Java Data Mining (JDM) provides a standard
application programming interface (API) and design framework to
provide developers, application architects, data analysts, and business
analysts greater access to data mining technology.
3