Python Natural Language Processing

Disadvantages of k-means clustering

These are the disadvantages of k-means clustering:

• Initialization of the cluster center is a really crucial part. Suppose you have three clusters and you put two centroids in the same cluster and the other one in the last cluster. Somehow, k-means clustering minimizes the Euclidean distance for all the data points in the cluster and it will become stable, so actually, there are two centroids in one cluster and the third one has one centroid. In this case, you end up having only two clusters. This is called the local minimum problem in clustering.

This is the end of the unsupervised learning algorithms. Here, you have learned about the k-means clustering algorithm and developed the document classification ...