Building a Clustering Model with Spark - Machine Learning with Spark - page 276

Making predictions using a clustering
model
Using the trained K-means model is straightforward and similar to the other models we
have encountered so far, such as classification and regression. We can make a prediction
for a single Vector instance as follows:
val movie1 = movieVectors.first
val movieCluster = movieClusterModel.predict(movie1)
println(movieCluster)
We can also make predictions for multiple inputs by passing a RDD [Vector] to the
predict method of the model:
val predictions = movieClusterModel.predict(movieVectors)
println(predictions.take(10).mkString(","))
The resulting output is a cluster assignment for each data point:
0,0,1,1,2,1,0,1,1,1
Tip
Note that due to random initialization, the cluster assignments might change from one run
of the model to another, so your results might differ from those shown earlier. The cluster
ID themselves have no inherent meaning; they are simply arbitrarily labeled, starting from
0.