Fuzzy clustering is similar to K-Means but with a key difference:
- data points within the fuzzy clusters do not belong to a singular cluster
- instead, each data point has a coefficient for each cluster, representing the likelihood of the datapoint being part of that cluster
- the centroid of a cluster is the mean of all points, weighted by their degree of belonging to that cluster
Fuzzy clustering is K-Means Clustering with soft membership, because each point keeps a coefficient for every cluster.