Programmer Guide/Command Reference/EVAL/modclust - Revision history

Christian at 18:13, 21 April 2011

2011-04-21T18:13:40Z

← Older revision		Revision as of 20:13, 21 April 2011
Line 16:		Line 16:
	\|2 \|\|Complete Linkage (log. distances)		\|2 \|\|Complete Linkage (log. distances)
	\|}		\|}
	:;<var>min, max</var>:optional minimum (default=2) and maximum (default=N) number of clusters for BIC calculation; 2 <= ''min'' < ''max'' <= N		:;<var>min, max</var>:optional minimum (default=2) and maximum (default=N) number of clusters for BIC calculation; 2 ≤ ''min'' < ''max'' ≤ N
	:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= ''alpha'' <= 100		:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 ≤ ''alpha'' ≤ 100
	;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable'' ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).		;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable'' ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).
	:Format of the hierarchy table ''htable'': N rows, 3 columns		:Format of the hierarchy table ''htable'': N rows, 3 columns

193.171.195.48 at 10:22, 21 April 2011

2011-04-21T10:22:08Z

← Older revision		Revision as of 12:22, 21 April 2011
Line 19:		Line 19:
	:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= ''alpha'' <= 100		:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= ''alpha'' <= 100
	;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable'' ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).		;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable'' ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).
	::hierarchy table ''htable'': N rows, 3 columns		:Format of the hierarchy table ''htable'': N rows, 3 columns
	::{\|class="einrahmen"		::{\|class="einrahmen"
	\|column 0 \|\|index of min. row (from)		\|column 0 \|\|index of min. row (from)
Line 27:		Line 27:
	\|column 2 \|\|agglomeration cost (distance)		\|column 2 \|\|agglomeration cost (distance)
	\|}		\|}
	::BIC table ''bictable'': ''max''-''min''+1 rows, 3 columns		:Format of the BIC table ''bictable'': ''max''-''min''+1 rows, 3 columns
	::{\|class="einrahmen"		::{\|class="einrahmen"
	\|column 0 \|\|number of clusters		\|column 0 \|\|number of clusters

193.171.195.48 at 05:32, 21 April 2011

2011-04-21T05:32:22Z

← Older revision		Revision as of 07:32, 21 April 2011
Line 49:		Line 49:
	:*otherwise: index vector containing the indices of the data vectors associated with cluster ''iclust''		:*otherwise: index vector containing the indices of the data vectors associated with cluster ''iclust''
	----		----
	;See also: [[../haclust\|haclust]], [[../em\|em]], [../density\|density]], [[../svd\|svd]]		;See also: [[../haclust\|haclust]], [[../em\|em]], [[../density\|density]], [[../svd\|svd]]

	[[../#Functions\|<function list>]]		[[../#Functions\|<function list>]]

193.171.195.48 at 05:31, 21 April 2011

2011-04-21T05:31:52Z

← Older revision		Revision as of 07:31, 21 April 2011
Line 46:		Line 46:
	:;<var>x</var>:input data matrix (NxM matrix, see ''Usage 1'')		:;<var>x</var>:input data matrix (NxM matrix, see ''Usage 1'')
	;Result 3:		;Result 3:
	:*if ''x'' is supplied: ~~data~~ matrix of all data vectors associated with the cluster ''iclust''		:*if ''x'' is supplied: matrix of all data vectors (rows of ''x'') associated with the cluster ''iclust''
	:*otherwise: index vector containing the indices of the data vectors associated with cluster ''iclust''		:*otherwise: index vector containing the indices of the data vectors associated with cluster ''iclust''
	----		----

193.171.195.48 at 05:30, 21 April 2011

2011-04-21T05:30:51Z

← Older revision		Revision as of 07:30, 21 April 2011
Line 18:		Line 18:
	:;<var>min, max</var>:optional minimum (default=2) and maximum (default=N) number of clusters for BIC calculation; 2 <= ''min'' < ''max'' <= N		:;<var>min, max</var>:optional minimum (default=2) and maximum (default=N) number of clusters for BIC calculation; 2 <= ''min'' < ''max'' <= N
	:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= ''alpha'' <= 100		:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= ''alpha'' <= 100
	;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).		;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable'' ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).
	::hierarchy table ''htable'': N rows, 3 columns		::hierarchy table ''htable'': N rows, 3 columns
	::{\|class="einrahmen"		::{\|class="einrahmen"

193.171.195.48 at 05:30, 21 April 2011

2011-04-21T05:30:33Z

193.171.195.48: Created page with '{{DISPLAYTITLE:{{SUBPAGENAME}}}} Model based agglomerative clustering. A hierarchy and BIC (Bayesian Information Criterion) values are calculated for given data vectors. ---- ;Us…'

2011-04-21T05:29:05Z

Created page with '{{DISPLAYTITLE:{{SUBPAGENAME}}}} Model based agglomerative clustering. A hierarchy and BIC (Bayesian Information Criterion) values are calculated for given data vectors. ---- ;Us…'

New page

{{DISPLAYTITLE:{{SUBPAGENAME}}}}
Model based agglomerative clustering. A hierarchy and BIC (Bayesian Information Criterion) values are calculated for given data vectors.
----
;Usage 1:<code>modclust(<var>x</var>, <var>htable</var>, <var>bictable</var>, <var>mflag</var> {, <var>min</var>, <var>max</var>, <var>alpha</var>})</code>
:;<var>x</var>:data matrix NxM, one data vector with length M per row
:;<var>htable</var>:hierarchy table (reference used for output)
:;<var>bictable</var>:BIC table (reference used for output)
:;<var>mflag</var>:method for distance and BIC calculation
::{|class="einrahmen"
!''mflag'' !! method
|-
|0 ||Single Linkage
|-
|1 ||Complete Linkage (linaer distances)
|-
|2 ||Complete Linkage (log. distances)
|}
:;<var>min, max</var>:optional minimum (default=2) and maximum (default=N) number of clusters for BIC calculation; 2 <= ''min'' < ''max'' <= N
:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= alpha <= 100
;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).
::hierarchy table ''htable'': N rows, 3 columns
::{|class="einrahmen"
|column 0 ||index of min. row (from)
|column 1 ||index of min. column (to)
|column 2 ||agglomeration cost (distance)
|}
::BIC table ''bictable'': ''max''-''min''+1 rows, 3 columns
::{|class="einrahmen"
|column 0 ||number of clusters
|column 1 ||log. likelihood
|column 2 ||BIC
|}
----
;Usage 2:<code>modclust(<var>htable</var>, <var>nclust</var>)</code>
:;<var>htable</var>:cluster hierarchy table (Nx3 matrix, see '''Usage 1''').
:;<var>nclust</var>:number of clusters
;Result 2:The created partition table ''p'', which is a vector with N elements containing the group indices. The value ''p''[i] (i=0..N) is the index of the cluster containing the data vector i.
----
;Usage 3:<code>modclust(<var>ptable</var>, <var>iclust</var> { <var>x</var>})</code>
:;<var>ptable</var>:partition table (Nx1 matrix, see '''Usage 2''').
:;<var>nclust</var>:index of cluster to be extracted
:;<var>x</var>:input data matrix (NxM matrix, see ''Usage 1'')
;Result 3:
:*if ''x'' is supplied: data matrix of all data vectors associated with the cluster ''iclust''
:*otherwise: index vector containing the indices of the data vectors associated with cluster ''iclust''
----
;See also: [[../haclust|haclust]], [[../em|em]], [../density|density]], [[../svd|svd]]

[[../#Functions|<function list>]]

← Older revision		Revision as of 07:30, 21 April 2011
Line 17:		Line 17:
	\|}		\|}
	:;<var>min, max</var>:optional minimum (default=2) and maximum (default=N) number of clusters for BIC calculation; 2 <= ''min'' < ''max'' <= N		:;<var>min, max</var>:optional minimum (default=2) and maximum (default=N) number of clusters for BIC calculation; 2 <= ''min'' < ''max'' <= N
	:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= alpha <= 100		:;<var>alpha</var>:optional factor for BIC calculation (default=1); 0 <= ''alpha'' <= 100
	;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).		;Result 1: On return the hierarchy information is stored in ''htable'' (Nx3 matrix) and the BIC values are stored in ''bictable ((''max''-''min''+1)x3 matrix). The return value ''ibest'' is the index of the BIC table entry with the highest BIC value (<code>''ibtest''=imax(''bictable''[*,2]</code>).
	::hierarchy table ''htable'': N rows, 3 columns		::hierarchy table ''htable'': N rows, 3 columns
	::{\|class="einrahmen"		::{\|class="einrahmen"
	\|column 0 \|\|index of min. row (from)		\|column 0 \|\|index of min. row (from)
			\|-
	\|column 1 \|\|index of min. column (to)		\|column 1 \|\|index of min. column (to)
			\|-
	\|column 2 \|\|agglomeration cost (distance)		\|column 2 \|\|agglomeration cost (distance)
	\|}		\|}
Line 28:		Line 30:
	::{\|class="einrahmen"		::{\|class="einrahmen"
	\|column 0 \|\|number of clusters		\|column 0 \|\|number of clusters
			\|-
	\|column 1 \|\|log. likelihood		\|column 1 \|\|log. likelihood
			\|-
	\|column 2 \|\|BIC		\|column 2 \|\|BIC
	\|}		\|}
Line 35:		Line 39:
	:;<var>htable</var>:cluster hierarchy table (Nx3 matrix, see '''Usage 1''').		:;<var>htable</var>:cluster hierarchy table (Nx3 matrix, see '''Usage 1''').
	:;<var>nclust</var>:number of clusters		:;<var>nclust</var>:number of clusters
	;Result 2:The created partition table ''p'', which is a vector with N elements containing the group indices. The value ''p''[i] (i=0..N) is the index of the cluster containing the data vector i.		;Result 2:The created partition table ''ptable'', which is a vector with N elements containing the group indices. The value ''ptable''[i] (i=0..N) is the index of the cluster containing the data vector i.
	----		----
	;Usage 3:<code>modclust(<var>ptable</var>, <var>iclust</var> { <var>x</var>})</code>		;Usage 3:<code>modclust(<var>ptable</var>, <var>iclust</var> { <var>x</var>})</code>