Q1-Compare K-means and K-medoids algorithms. List down the main differences betw

Question

Q1-Compare K-means and K-medoids algorithms. List down the main differences between these two algorithms?

Q2-Cluster the following eight points into three clusters using K means clustering algorithm and use Euclidean distance.

              A1=(2,10), A2=(2,5), A3=(8,4),   A4=(5,8),

              A5=(7,5),   A6=(6,4),   A7=(1,2),   A8=(4,9).

a)      Create distance matrix by calculating Euclidean distance between each pair of points. (0.5 mark)

b)      Suppose that the initial centers of each cluster are A1, A4 and A7. Run the k-means algorithm for once only and show:

                  i.            The new clusters (i.e. the examples belonging to each cluster) (1 mark)

                ii.            The centers of the new clusters (0.5 mark)

Show all your work.

Explanation / Answer

Q1-Compare K-means and K-medoids algorithms. List down the main differences between these two algorithms?

Answer)

K-means and K-medoids are both clustering algorithms means for grouping a list of objects to each other such that like objects are in the same group. Cluster analysis and grouping algorithms are many in such a way different approaches are there for the clustering of the similar objects into groups.

The k-means algorithm for clustering is used to partition and group n observation into k clusters such that every observation occurs in the cluster with the closest means. This results in partitioning and grouping of the set of data. Upon converging multiple times on the same group the k-means result no longer change for a data set and is regarded as final.

The K-medoids algorithm is used to cluster the data set with the use of and related to the k-means algorithm and the medoidshift algorithm. Both the algorithms in the question has the purpose of minimizing the distance between the data points which are in a group or cluster and a point which is in a center of a cluster.

However there are differences such as as not in the K-means algorithm, in the K-medoids algorithm the centers are chosen as the datapoints. K-means does minimize the total squared error but the K-medoids algorithm minimizes the sum of differences between the data points which are in a group or cluster and a point which is in a center of a cluster. K-medoids is more robust than K-means as it is more robust to noise.

Related Questions

Navigate

• Browse (All)
• Browse Q
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.