!!This is file autoread.txt. If you do not want this file to be loaded automatically, simply delete it or 
rename in your directory!!

To make the analysis of data table, it is necessary to do the following steps:

1.Open data table file  (File|Open). If this file does not exist, choose File|New (for the convenience, 
it is possible to use correspondent pictograms).

Given file must be a text file and written in following format, for instance: (text between 
symbols # # is a commentary ignored by program)

   22    # number of objects (cases); in the given version - no more than 1000#
   16    #number of variables; in the given version - no more than 50# 
    N   #code of unknown value#

 #types of variables: 1 - quantitative; 0(M) - qualitative, M =number of values# 
1 0(4) 0(4) 0(4) 0(4) 0(4) 0(4) 0(4) 0(4) 0(4) 0(4) 0(4) 1 1 1 1

 #names of  variables:#
Y X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15

# data table; numbers must be separated by the space or comma; decimal values are separated by dot#
  0  1  2  3  2  4  3  4  1  3  1  3  3  2  4  2
  2  2  2  4  4  4  3  4  3  2  3  3  0  3  4  4
  0  4  4  4  3  4  3  2  3  3  2  3  0  2  5  4
  4  1  1  1  1  1  3  2  3  3  1  1  7  1  3  0
 14 3  4  3  2  3  3  4  1  4  1  1  3  1  4  3
15  3  3  4  3  3  3  4  1  1  1  4  3  0  5  3



You may edit this file, using built-in editor

2. Assign parameters of algorithm  (Run|Parameters)

Recursive complexity:
Defines an amount of variables, falling into the combination considered by the algorithm in each step 
(search depth). When increasing this parameter, increase both a quality of decision, and time and 
required memory.Integer part of the parameter assigns a guarantee size (r) of a combination of variables.
Fractional part of the parameter assigns a share of the most informative variables (for which the 
expected error is minimal), selected for combinations of amount r+1, r+2 and etc (from variables, 
chosen on the preceding stage).

Homogeneity parameter:
The node does not need further branching if the objects in the node are uniform. The objects are 
considered as being uniform, if the relative variance for predicting variable (or percentage of recognition 
error, in case of qualitative variable forecasting) is less than the given parameter.

MinObjInNode (minimum number of objects in the node):
Defines a minimum possible number of objects in the node which needs further branching 
(i.e. if in the node has less objects than it is specified, no branching is produced from this node).

VariantsNumber: 
Desired number of variants of decision tree. The variants are built by the way of consequent excluding
the most informative (corresponding to the root node) variable in the preceding variant. 
If test sample is given, the variants are used for final decision design by voting procedure taking
into account the competence of each tree variant. The competence estimations are inversly
proportional to crossvalidate error for each tree.

FeatureToPredict: 
Number of predicting variable in data table;

FoldsCrossValidate:
Parameter L in the method of L-fold cross validation. In this method, the initial sample is divided 
randomly on L parts of approximately equal size. Then each part alternately serves as test sample, 
whereas other parts are united into learning sample. As a factor of quality of a method, the averaged 
test samples error is calculated. In the case of regression tree, the error (standard deviation) is 
normalized to the standard deviation of the initial sample.

3. To Start program of  decision tree design (Run|Grow Tree)

4. Results are automatically saved in file 'output.txt'. 

