 1. Title: Waveform Database Generator (written in C)

 2. Source:
 (a) Breiman,L., Friedman,J.H., Olshen,R.A., & Stone,C.J. (1984).
 Classification and Regression Trees. Wadsworth International
 Group: Belmont, California. (see pages 4349).
 (b) Donor: David Aha
 (c) Date: 11/10/1988

 3. Past Usage:
 1. CART book (above):
  Optimal Bayes classification rate: 86% accuracy
  CART decision tree algorithm: 72%
  Nearest Neighbor Algorithm: 78%
  300 training and 5000 test instances

 4. Relevant Information:
  3 classes of waves
  21 attributes, all of which include noise
  See the book for details (4955, 169)

 5. Number of Instances: chosen by user

 6. Number of Attributes:
  21 attributes with continuous values between 0 and 6

 7. Attribute Information:
  Each class is generated from a combination of 2 of 3 "base" waves
  Each instance is generated f added noise (mean 0, variance 1) in
 each attribute
  See the book for details (4955, 169)

 8. Missing Attribute Values: none

 9. Class Distribution: 33% for each of 3 classes


0,1,2
X00 : continuous.
X01 : continuous.
X02 : continuous.
X03 : continuous.
X04 : continuous.
X05 : continuous.
X06 : continuous.
X07 : continuous.
X08 : continuous.
X09 : continuous.
X10 : continuous.
X11 : continuous.
X12 : continuous.
X13 : continuous.
X14 : continuous.
X15 : continuous.
X16 : continuous.
X17 : continuous.
X18 : continuous.
X19 : continuous.
X20 : continuous.