PISAKNN uses KNN (k nearest neighbours) distribution-free
multivariate discrimination to classify objects into two classes. The classes are seeded by
supplying two files which contain the indices of objects typical to the class in question
(5, approximately
equal numbers of each). Each object then propagates its class to the other objects on the basis of which class
of the 2k
nearest neighbours (in the parameter space of the PISAPEAK results) of each of the unclassified
objects is most common. This procedure is iterated until all objects are assigned and have a stable class
or until a maximum number of iterations is exceeded. The results of the discrimination are written
into two output files, one for each class.
CLASS1 = FILENAME (Write)
Name of a file to contain the indices of
the objects selected for membership of class 1. [CLASS1.DAT]
CLASS2 = FILENAME (Write)
Name of a file to contain the indices of the objects selected for membership of class 2.
[CLASS2.DAT]
ELLIP = _LOGICAL (Read)
If ‘true’ then the ellipticities are used in the analysis.
If ‘false’ then they are excluded. Using ellipticities may increase the weighting of some
(small) round galaxies as stars. [TRUE]
K = _INTEGER (Read)
The number of nearest
neighbours about the current values which are to be used in classifying an object. The
class used is the most frequently encountered in this range of objects. If classes 1 and 2 are
equally frequent then the object classification is not changed. [1]
NITER = _INTEGER
(Read)
The maximum number of iterations allowed to classify and reclassify objects. [10]
PEAKDATA = FILENAME (Read)
Name of a file containing the results of the PISAPEAK
parameter transformation. This file must contain at least five columns which have the
values:
- object index
- radius ratio
- intensity-peak ratio
- ellipticity
- absolute value of intensity weighted cross moment
in that order. [PISAPEAK.DAT]
SEED1 = FILENAME (Read)
Name of a file containing
the indices of the objects to seed class1. The file can contain any number of columns but
must have the object indices in column one. [SEED1.DAT]
SEED2 = FILENAME (Read)
Name of a file containing the indices of the objects to seed class2. The file can contain
any number of columns but must have the object indices in column one. [SEED2.DAT]