This paper shows an interesting and novel method for classification of lava fountains at Etna based on a machine-learning technique. The scientific approach is valid and potentially applicable to other volcanoes. I appreciate the suggestions of the reviewers because they help in improving the quality of the paper. In particular, I agree with reviewer #2 about the organization of section 3. Adding a brief introduction about the k-means method and a short descriptions of the used concepts (such as centroid, etc.) will simplify the reading.
Technical comment: At page 10 of the manuscript with track-changes, attached to the reply of reviewer 2, at Line 188, "max{Sila,Ci,j} <= Sa,Ci". In the term on the right, probably do you mean "<= Sila,Ci"?

I wish to thank the reviewers for the constructive suggestions that improved the readability of the manuscript. The authors answered satisfactorily to all the comments of the reviewers.