Identifies specimens using a random classification forest (R.W. Payne).
Options
PRINT = string tokens |
Controls printed output (identification); default * i.e. none |
|---|---|
IDENTIFICATION = scalar or variate |
Saves the identification of each specimen |
VOTES = matrix |
Saves numbers of the terminal nodes reached by the specimens |
SAVE = pointers |
Save structure from BCFOREST providing information about the random forest |
arameters
X = variates or factors |
Explanatory variables |
|---|---|
VALUES = scalars, variates or texts |
Values to use for the explanatory variables; if these are unset for any variable, its existing values are used |
Description
BCFIDENTIFY identifies specimens using a random classification forest, constructed by the BCFOREST procedure. The SAVE parameter can be set to a pointer, saved using the SAVE option of BCFOREST, containing the necessary information about the forest. Alternatively, if you do not set SAVE, the identification will be made using the forest most recently constructed by BCFOREST.
The characteristics of the specimens can be specified in the variates or factors listed by the X parameter. These must have identical names (and levels) to those used originally to construct the tree. You can use the VALUES parameter to supply new values, if those stored in any of the variates or factors are unsuitable.
The PRINT option controls printed output, with settings:
identification |
to print the identifications obtained using the tree. |
|---|
By default nothing is printed.
The IDENTIFICATION option allows you to save the identifications (in a scalar or variate according to whether there is one or several specimens); a missing value is given if there is no clear result (i.e. more than one group possible) for the specimen concerned. The VOTES option can save a specimens-by-groups matrix with the votes given by the forest for each of the groups with each specimen.
Options: PRINT, IDENTIFICATION, VOTES, SAVE.
Parameters: X, VALUES.
Action with RESTRICT
Restrictions are ignored.
See also
Procedures: BCFOREST, BCFDISPLAY.
Commands for: Multivariate and cluster analysis.
Example
CAPTION 'BCFOREST example',\
!t('Random classification forest for automobile data',\
'from UCI Machine Learning Repository',\
'http://archive.ics.uci.edu/ml/datasets/Automobile');\
STYLE=meta,plain
SPLOAD FILE='%gendir%/examples/Automobile.gsh'
BCFOREST [GROUPS=symboling; NTREES=8; NXTRY=10; NUNITSTRY=75; SEED=197883]\
normalized_losses,make,fuel_type,aspiration,number_doors,\
body_style,drive_wheels,engine_location,wheel_base,\
length,width,height,curb_weight,engine_type,number_cylinders,\
engine_size,fuel_system,bore,stroke,compression_ratio,\
horsepower,peak_rpm,city_mpg,highway_mpg,price
BCFIDENTIFY [PRINT=identification]\
normalized_losses,make,fuel_type,aspiration,number_doors,\
body_style,drive_wheels,engine_location,wheel_base,\
length,width,height,curb_weight,engine_type,number_cylinders,\
engine_size,fuel_system,bore,stroke,compression_ratio,\
horsepower,peak_rpm,city_mpg,highway_mpg,price