K-Means Clustering (Material from cs412, uiuc) - The Lack Thereof
K-Means Clustering (Material from cs412, uiuc) - The Lack Thereof
K-Means Clustering (Material from cs412, uiuc) - The Lack Thereof
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
What Is the Problem of the K-<strong>Means</strong> Method<br />
●<br />
<strong>The</strong> k-means algorithm is sensitive to outliers !<br />
●<br />
Since an object with an extremely large value may substantially<br />
distort the distribution of the data.<br />
●<br />
K-Medoids: Instead of taking the mean value of the object in a cluster<br />
as a reference point, medoids can be used, which is the most<br />
centrally located object in a cluster.<br />
10<br />
9<br />
8<br />
7<br />
6<br />
5<br />
4<br />
3<br />
2<br />
1<br />
0<br />
0 1 2 3 4 5 6 7 8 9 10<br />
10<br />
9<br />
8<br />
7<br />
6<br />
5<br />
4<br />
3<br />
2<br />
1<br />
0<br />
0 1 2 3 4 5 6 7 8 9 10<br />
July 22, 2010 Data Mining: Concepts and Techniques 6
Change language