I am working on finding optimal number of clusters for IoT dynamic multivariate data. Can you please suggest which method/technique best suites for my problem.
Well, finding optimal number of clusters is an NP-Hard problem if number of classes are not known. Best way to find the solution of NP-Hard problem is to use evolutionary optimization techniques like Particle Swarm Optimization and Antlion Optimizer. Moreover, you can also infuse the process of feature selection for cluster selection.