DDE-1/Clustering.md

@@ -64,6 +64,11 @@ In order to group the observations, we need a measure of distance. If it is set,
 
 The third method is what is usually preferred. In any case, selecting the right number of cluster is not easy. Once the tree is constructed, we search for large gaps in the merged sub-groups, indicating the dissimilarities between the merged groups. The gap detection process can be done by the user. Alternatively, we can follow the [Bayesian treatment](https://www2.stat.duke.edu/~kheller/bhcnew.pdf). 
 
+### Density based clustering
 
+...
+
+### Concept of similarity in machine learning
 
+...