After a clustering model is trained, before you start the prediction process, make sure you merge features such as ID, time, and other related features to better understand the relationships between different fields. The ID column is a unique identifier to your data and should always be added to the merge column field.