| LED 8 names file.
| Created by YeoGirl Yun
| Date : 5/14/94
| The .all file is created using the following command :
|  led-creator 3200 led7.all 239418 10
| The .data file is the first 200, and the test is the rest.
|
| This data file is created by using AHA's led creator program.
| Here is the relevant document from AHA
|
|------------------------------------------------------------------------------
|1. Title of Database: LED display domain
|
|2. Sources:
|   (a) Breiman,L., Friedman,J.H., Olshen,R.A., & Stone,C.J. (1984). 
|       Classification and Regression Trees.  Wadsworth International
|       Group: Belmont, California.  (see pages 43-49).
|   (b) Donor: David Aha 
|   (c) Date: 11/10/1988
|
|3. Past Usage: (many)
|     1. CART book (above):
|        -- Optimal Bayes classification rate: 74%
|        -- CART decision tree algorithm: 71% (resubstitution estimate)
|        -- Nearest Neighbor Algorithm: 71%
|           -- 200 training and 5000 test instances
|     
|     2. Quinlan,J.R. (1987). Simplifying Decision Trees.  In International
|        Journal of Man-Machine Studies (to appear).
|        -- C4 decision tree algorithm: 72.6% (using pessimistic pruning)
|           -- 2000 training and 500 test instances
|     3. Tan,M. & Eshelman,L. (1988). Using Weighted Networks to Represent
|        Classification Knowledge in Noisy Domains.  In Proceedings of the
|        5th International Conference on Machine Learning, 121-134, Ann
|        Arbor, Michigan: Morgan Kaufmann.  
|        -- IWN system: 73.3% (using the And-OR classification algorithm)
|           -- 400 training and 500 test cases
|
|4. Relevant Information Paragraph:
|     This simple domain contains 7 Boolean attributes and 10 concepts,
|     the set of decimal digits.  Recall that LED displays contain 7
|     light-emitting diodes -- hence the reason for 7 attributes.  The
|     problem would be easy if not for the introduction of noise.  In
|     this case, each attribute value has the 10% probability of having
|     its value inverted.  
|
|     It's valuable to know the optimal Bayes rate for these databases.
|     In this case, the misclassification rate is 26% (74% classification
|     accuracy).
|        
|5. Number of Instances: chosen by the user.
|
|6. Number of Attributes: 7 (all Boolean-valued)
|
|7. Attribute Information:
|   -- All attribute values are either 0 or 1, according to whether
|      the corresponding light is on or not for the decimal digit.
|   -- Each attribute (excluding the class attribute, which is an
|      integer ranging between 0 and 9 inclusive) has a 10% percent
|      chance of being inverted.
|
|8. Missing Attribute Values: None
|
|9. Class Distribution: 10% (Theoretical)
|   -- Each concept (digit) has the same theoretical probability
|      distribution.  The program randomly selects the attribute.
|
|------------------------------------------------------------------------------

0,1,2,3,4,5,6,7,8,9. | classes

attribute#1 : 0,1.
attribute#2 : 0,1.
attribute#3 : 0,1.
attribute#4 : 0,1.
attribute#5 : 0,1.
attribute#6 : 0,1.
attribute#7 : 0,1.