Scribd
Upload
57 views2 pages

Lecture32 K-Means Clustering Exercise

The document discusses the k-means clustering algorithm applied to a 1-dimensional dataset with 10 data points from 1 to 10. It shows the results of running k-means with k=2 and different initial seed sets. With an initial seed of {1,2}, it takes 3 iterations to converge, while with {2,9} the algorithm converges in 1 iteration, demonstrating that the choice of initial seeds can impact the quality and speed of convergence.

Uploaded by

Pavan Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
57 views2 pages

Lecture32 K-Means Clustering Exercise

The document discusses the k-means clustering algorithm applied to a 1-dimensional dataset with 10 data points from 1 to 10. It shows the results of running k-means with k=2 and different initial seed sets. With an initial seed of {1,2}, it takes 3 iterations to converge, while with {2,9} the algorithm converges in 1 iteration, demonstrating that the choice of initial seeds can impact the quality and speed of convergence.

Uploaded by

Pavan Kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Nitte Meenakshi Institute of Technology

Department of Computer Science & Engineering


Course Coordinator: Dr.Vani V

Clustering: K-Means
Exercise:
1. Consider the 1-dimensional data set with 10 data points {1, 2, 3, . . . 10}. Show three
iterations of the k-means algorithms when k = 2, and the random seeds are initialized to
{1,2}. Repeat the problem with random seeds {2,9}. How did the different choice of the
seed set affect the quality of the results?
Use Manhattan |m1-xi| |m2-xi| i=1 to 10

Solution:
xi m1(1) m2(2) MANHATTAN
1 0 1
2 1 0
3 2 1
4 3 2
5 4 3
6 5 4
7 6 5
8 7 6
9 8 7
10 9 8

INITIAL: C1={1} C2={2,3,4,5,6,7,8,9,10} 6

dp m1(1) m2(6)
1 0 5
2 1 4
3 2 3
4 3 2
5 4 1
6 5 0
7 6 1
8 7 2
9 8 3
10 9 4
ITERATION
1 C1={1,2,3} C2={4,5,6,7,8,9,10} 7
dp m1(2) m2(7)
1 1 6
2 0 5
3 1 4
4 2 3
5 3 2
6 4 1
7 5 0
8 6 1
9 7 2
10 8 3

ITERATION
2: C1={1,2,3,4} 2.5 C2={5,6,7,8,9,10} 7.5
dp m1(2.5) m2(7.5)
1 1.5 6.5
2 0.5 5.5
3 0.5 4.5
4 1.5 3.5
5 2.5 2.5
6 3.5 1.5
7 4.5 0.5
8 5.5 0.5
9 6.5 1.5
10 7.5 2.5

ITERATION 3: c1={1,2,3,4,5}
c2=(6,7,8,9,10}

Ö With the initial seed as {2,9}; the quality of the algorithm is improvised, and the
algorithm converges in the first iteration itself.
Ö Initial:C1(m1=2) = {1,2,3,4,5}
C2(m2=9) = {6,7,8,9,10}
Ö Iteration 1: C1(m1=3) = {1,2,3,4,5}
C2(m2=8) = {6,7,8,9,10}

You might also like