You run K-means clustering on your dataset with K = 5 and get some result. Your coworker takes the same dataset, runs K-means clustering with K = 5 on her computer as well but gets a different result. How can this be possible, and what is one way to decide whose clustering result is better?