The correct answer is:

• Distance between the points and a cluster centre

K-means clustering is a popular unsupervised learning algorithm that partitions data points into clusters. It does so by minimizing the sum of squared distances between each data point and the centroid of the cluster it belongs to.

Would you like further details or have any questions?

Here are 5 related questions to consider:

1. How does K-means clustering initialize the cluster centers?
2. What are some common methods to choose the value of K in K-means?
3. How does the K-means algorithm handle non-spherical clusters?
4. What is the difference between K-means and K-medoids?
5. How does the K-means++ algorithm improve over the traditional K-means?

Tip: K-means is sensitive to the initial placement of centroids, which can affect the final clusters formed. Using techniques like K-means++ helps improve this initialization step.