Fuzzy Modeling and Genetic Algorithms for Data Mining and Exploration

1.4 Information and Knowledge Discovery

■

17

■

The Chief Information Officer of a large bank needed a way to con-

trol cost overruns on a rising number of high-technology projects in

the information systems group. Using a large database of failed or

inadequately delivered projects, the bank ranked each project with

a value of from 0 to 1 according to its perceived value (0 as a failed

project with no value to 1 as a completely successful project). They

also created fields to indicate the type of project, its perceived level

of complexity, and its degree of dependence on advanced technol-

ogy. A supervised data mining analysis discovered a clear relationship

between low user satisfaction (as well as project failure) and a set of

important project characteristics ...