Fuzzy k-means
Fuzzy k-means clustering is similar to k-means clustering but unlike k-means the clusters can be overlapping. A single point can belong to more than 1 cluster. Fuzzy k-means we need to estimate the number of clusters k and additionally the fuzziness factor m.
Deciding the fuzzy factor
The fuzziness factor determines the degree of overlap in the clusters. If the fuzziness factor is 1 then fuzzy k-means behaves like k-means, as the fuzziness factor increases we see increased overlap in the clusters.
Get Learning Apache Mahout now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
