Scribd
Upload
95 views

K-Means Clustering Numerical Example

Uploaded by

Aakash Bhat
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
95 views

K-Means Clustering Numerical Example

Uploaded by

Aakash Bhat
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

K-means

Clusterirg
Data Point

2 2

2 5
6

|2
13

cetrids
Randomly select two
C; = (2, 3)
- (6, )
Dut Euclidean dis tance
Now find
D=

Data Point Cluster


2 3
2

2. 2 3
Je-2+ (s3)
3 5 5123

5 Jo 5
12-04

13 1345
Euclidean
Eudlidean to distane to cluste
Data Point dittane

|.4| 860

72|
2 3 2
2
593 J.4)
2
3
1:21
6 9 2
|2.0 4 5
5 lo |2 2-.
565
13
l3-45

mea n
taki ny the
Vbdate Centids
each clus tey
data poins
and 2
data þoint
are

COts ter |
Mean (+2), (2+3)&)
2 2

Nens Centoidl G =|5, 2:5

data poi nt Ye 3, 4 5, 6

Mean (5+6+ lo +n) 9


(8++12 +13)

-] 8, lo 5)
2
Nous find out Nen arsgned clusters
the nens Centn'ds
besed upon

Data Point
15 25 Jo -5
0:707
2
0707
2 3
2
6519
8
3

I2:74
5
13 14.15

ertoid G, [S, 25)


Eucli di nean olis tance for
for Data Point I= J(s-°r (a5-2)=
t= (65-2*+ (25-3) oo7

(5- )+ -8)-
(a5-8) -st(-s9
3=

(:5-9) )ks
5-(15-)+ (215- )=J((n9 J72- 25+ o s
IC25 =12:79

2 14 I5
Euclidan diztane tor centid (8, lo5]

J(8-I) +(io5- 2) = 49 + 722s

W(3-2) +(o5-3) (9+(19*


92:5-60

8) (a5)
(8-5 )+ (oS -1s-25-30

(2)+(us)
44 2 s
25

(8- jo) (1o:5-I2)2)i(f

Vo-n)+(ws- 1s).e(aj.

Euclidean Eu cei dean 3.90


Data Point dstance for diatanu for
cleetr
070 7
2
960
2 2
2 2
30
3 2S 2
6
2
|2:74 25
|2
5 I4.1S 3.90
13
6
Calelated fo
For thìs da taset We have
terations.

both the
And
cluots ae Aame So we amigna

Data toi nts I, 2

cluster 2 Date Poirg 3,4,s,6

the value 6
both Ccnes

amijndl clusters are olfener then

tuke
ansipnel clete
data þol nts ntil thy con ve

cluste

You might also like