6) Clustering Flashcards
(4 cards)
1
Q
What is the basic idea of the K-Means clustering algorithm
A
2
Q
What is “hard clustering” in the context of K-Means, and why might it not be sufficient
A
K-Means is an example of hard clustering, which means that each data point is assigned to exactly one cluster — there is no overlap. However, in practice, especially when clusters are not well-separated, this can be limiting. Some data points might lie between clusters, and it would be more realistic to allow points to belong to multiple clusters with certain probabilities.
3
Q
How are cluster probabilities assigned in Gaussian Mixture Model (GMM)
A
4
Q
What is the Expectation-Maximisation (EM) algorithm
A
