An Algorithmic Introduction to Clustering
This work provides a more cohesive understanding of clustering for researchers and practitioners, but it is largely incremental as it builds on existing algorithms.
The paper tackles the problem of unifying clustering algorithms by identifying relationships between five methods, presenting results in a cleaner and simpler way, with a novel interpretation linking DBSCAN to Mean shift as a climbing procedure.
This paper tries to present a more unified view of clustering, by identifying the relationships between five different clustering algorithms. Some of the results are not new, but they are presented in a cleaner, simpler and more concise way. To the best of my knowledge, the interpretation of DBSCAN as a climbing procedure, which introduces a theoretical connection between DBSCAN and Mean shift, is a novel result.