DATA MINING
Desktop Survival Guide by Graham Williams |
|||||
Parametric Techniques: Analysis techniques which assume a model for the data to be analysed. The task of analysis is to select an appropriate model and then to estimate the parameters of the model.
Parallel Coordinates: A multidimensional visualisation technique where the dimensions are represented as a series of parallel lines and individuals are represetned as paths through the parallel coordinates. The idea was first presented by mailto:aiisreal@math.tau.ac.ilAlfred Inselberg at the University of Illinois in 1959 and he has been developing the idea since. Wegman has also written on Inselberg's work. See the Journal of Computational Statistics, January 1999 for example, for recent significant developments in Statistics.
Persistence: A general term for mechanisms that save values from a program's execution space so that they can be used in a later execution; i.e. by making the values "persist" from one execution to the next. See also Orthogonal Persistence.
Predictor: The statistical term for an input variable.
p-value: The probability that a ``discovery'' is pure chance. A p-value of 0.05 indicates that there is a 5% chance that the discovery is purely by chance, or conversely, that the discovery is 95% likely to be valid.
Copyright © 2004-2006 Graham.Williams@togaware.com Support further development through the purchase of the PDF version of the book.