Statistica 6 : joining (tree clustering), two-way joining, k-means clustering (k-).
( ). . MS Excel. .
k-means clustering (k-). , .
k- , .
, .
. ? , : (, , , , , ).
. ? , . .
, 2. . , . .
k- :
- ( );
- ;
- : ;
- ;
- 2, 3 , ( ) . .
, , , , , , 60 . , .
Cluster Analysis ( )
1. Statistica Cluster Analysis ( ), ( 16.1).
16.1
Claster Analysis ( ) ( 16.2).
|
|
16.2
2. , Open Data, ( 16.3).
16.3
3. Cars.sta, , . , , .
, , ( 16.4).
16.4 Cars.sta
, Cars.sta , , , .
22 . .
:
- PRICE ;
- ACCELE HANDLI ;
- MILAGE ( , ).
: , price
. .
, , ( , ).
, - , .
, , . . , . , , , , , .
, , , , . , .
, .
. , Statistica.
4. k-means (k-)
( 16.2) OK .
k-means ( 16.5).
16.5 k-means
5. . , .
Variables () : Select variable for the analysis ( ) ( 16.6).
16.6
, Select All ( ), OK.
6. Cluster (), Variables (). , Cases (). Variables (). , .
|
|
, , Cases.
7. Number of clusters ( ) , . 3.
, 3 .
8. Number of iterations ( ) , . , , 11.
9. Missing data (, - ). .
Initial cluster centers .
, 16.5.
10. , k-means Clustering .
11.
k-means Clustering
( 16.7).
16.7
: ,
, , , , :
Solution was obtained after 3 iterations (
3 ).
.
Analysis of variation ( ) .
Cluster Means&Euclidean Distances , ( ), .
Graph of means .
Descriptive Statistics for each clusters (, . .)
Save classifications and distances Statistica .
12. , . Member of each
cluster&distances.
3 , ( 16.8 16.10).
16.8
16.9
16.10
.
13. Cluster means&Euclidean distances. ,
( , 3 , ) ( 16.11).
16.11
, 0,969, , 1,876.
.
14. Graph of means ( ) ( 16.12).
16.12
k.
.
15. Variables () : Select variables for the analysis. , 16.13 ( 3 , ).
|
|
16.13 k-
16. , . Graph of means ( ),
( 16.14).
16.14
, . . ,
( 16.12).
. , .
, .
, , two-way joining, .
17. Joining (tree clustering), , ( 16.15), .
16.15