CL CLA - Classification - Clustering

PURPOSE

To find clusters of images in factor space (or a selected subspace) by using Didat's method of moving centers, and to apply hierarchical ascendant classification (HAC) to the resulting cluster centers.

SEE ALSO

CL HC	[Classification - Hierarchical Clustering]
CL HD	[Classification - Hierarchical clustering, calculate classes]
CL HE	[Classification - Hierarchical clustering, create docfiles]
CL KM	[Classification - K Means clustering ||]

USAGE

.OPERATION: CL CLA

.CORAN/PCA FILE PREFIX: COR_01_
[Enter name of the prefix of the image factor coordinate file (_IMC) containing your data. This file was created by 'CA S.

.CLUSTER OUTPUT FILE: CLU001
[Enter name for cluster file where cluster and classification parameters are placed for later reference. This file must NOT exist currently. The only current use for this file is in: 'CA SM ]

.NUMBER OF FACTORS: 6
[Enter number of factors to be used in the cluster search. Uses this many factors starting with factor: 1.]

.NUMBER OF ITERATIONS & CENTERS PER PARTITION: 5,8
[Enter number of iterations per partition, (default=5) and number of centers ("seeds") per partition (default=5).]

.NUMBER OF PARTITIONS: 3
[Enter number of partitions (default=4).]

.LEVEL FOR CLASS CUTOFF (0=NO CUTOFF): 2.4
[Enter the cutoff level at which the dendrogram will be cut. Classes having fewer occupants than this level will be removed. This number must be be in the range of the scaling of the dendrogram tree which was just reported.]

.DO YOU WANT A DENDROGRAM POSTSCRIPT PLOT (Y/N): Y
[If the answer is "Y" a dendrogram plot will be made.]

If the answer is "Y" the name of the output file will be solicited:

.ENTER POSTSCRIPT OUTPUT FILE: DEN001

.DO YOU WANT A DENDROGRAM DOC. FILE? (Y/N): Y
[Allows preparation of a dendrogram document file for hierarchical clustering.]

If the answer is "Y" the name of the document file will be solicited:

.DOCUMENT FILE: DOC001

NOTES

1. The RESULTS file contains the following information:
   A) Aggregation history:
   For each of the NP partitions, the NS seeds and sizes of clusters are listed. The non-empty clusters in the crossed partition are given by size and cummulative percentage. A total of NS**NP clusters are possible, but in practice, only 10% of these are non-empty.
   B) Description of hierarchy nodes:
   The nodes are numbered starting from the number of the highest cluster. For each node, the Senior, Junior, size, weight and hierarchy index are given. The hierarchy index is printed out as a histogram.
   C) Description of the classes:
   For each node the constituent classes are listed.
   D) Dendrogram:
   The class relationships are represented in the form of a dendrogram (tree structure). The lengths of the branches (in horizontal print direction) are proportional to the heirarchy indices.
   E) List of class members:
   The members are listed for each of the basic non-empty classes.
   F) List of class center coordinates:
   For each class, the NFAC coordinates of its center is listed. These are contained in the cluster file.
   G) Re-classification lookup table:
   Each cutoff point in the dendrogram, from right to left, defines a new classification scheme with the number of classes increasing by 1 each time. The table gives the new class memberships for any cutoff point selected.
   

2. The algorithm underlying this set of programs is described by Lebart et al., MULTIVARIATE DESCRIPTIVE STATISTICAL ANALYSIS, J. Wiley and Sons, New York 1984, Chapter V, p.109.

SUBROUTINES

SCLASSI, NOYAU, CHAVA, DEUCL, RGRI, RCLU, DENDRO, ARBRE, PARST, STABK, SHELK, COUPE, CLAST, RETIR, SEN3A, DENDRO2, DENLST

CALLER

UTIL1