I am currently working on a project where I need to segment the users into three categories: novice, intermediate and expert. But I have to cluster them using the user logs. Now the user logs contain scripts of some tool usage. Basically, I have to analyse how many times the user has successfully executed the tools, how many tools he has used so far and some other criteria. So in order to accomplish my task, how should I proceed? Do I need to use both NLP and clustering algorithms for clustering the existing user data and predicting the future data?
It would be really helpful for me if I can get more insight about how should I proceed on doing the task.
Thanks in advance.