Description

K-means is an unsupervised machine learning method to split data into K groups in which each observation belongs to the groups with the nearest mean hence can serve as a representative member of the group.

Expected Deliverable

A k-means pipeline operator. Details (the algorithm e.g. Lloyd's vs Hartigan-Wong, initialization options, additional metrics to be returned: SSB, SSW etc.) to be specified during spec'ing. 

Expected Teams Impacted

Server-Query