Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection

By Ricardo José Gabrielli Barreto Campello, Davoud Moulavi, Arthur Zimek, Jörg Sander. ACM Transactions on Knowledge Discovery from Data, 2015

853 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Epidemiology

View Paper: https://www.semanticscholar.org/paper/7e99ce2002650b45d291df90c64ad586833a3c2a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Arthur Zimek
2. Jörg Sander
3. Ricardo José Gabrielli Barreto Campello
4. Leland McInnes
5. John Healy
6. Hongzhi Wang
7. Mohamed Hammad
8. Mohamed Jaward Bah

In Epidemiology

1. John Healy
2. Leland McInnes
3. Mohamed Hammad
4. Mohamed Jaward Bah
5. Hongzhi Wang
6. Jörg Sander
7. Ricardo José Gabrielli Barreto Campello
8. Arthur Zimek

Citations in the corpus

• A Deep Neural Network for Unsupervised Anomaly Detection and Diagnosis in Multivariate Time Series Data (2018, methodology): “…techniques include distance/clustering methods (He, Xu, and Deng 2003; Hautama¨ki, Ka¨rkka¨ınen, and Fra¨nti 2004; Id´e, Papadimitriou, and Vlachos 2007; Campello et al. 2015), probabilistic methods (Chandola, Banerjee, and Kumar 2009), density estimation meth-ods (Manevitz and Yousef 2001),…”
• hdbscan: Hierarchical density based clustering (2017, methodology): “The library also includes support for Robust Single Linkage clustering (Chaudhuri et al. 2014), (Chaudhuri and Dasgupta 2010), GLOSH outlier detection (Campello et al. 2015), and tools for visualizing and exploring cluster structures.”
• Progress in Outlier Detection Techniques: A Survey (2019, methodology): “[86] technique, and they experimentally proved that themethod outperforms othermethods [8], [26] in terms of the running time and detection rate.”
• ContactPose: A Dataset of Grasps with Object Contact and Hand Pose (2020, neutral): “Organizing the grasps by clustering these aligned 3D joints (using L2 distance and HDBSCAN [7]) reveals the diversity of grasps captured in ContactPose (Figure 9).”
• Accelerated Hierarchical Density Based Clustering (2017, methodology): “As described in [7] and [8] the HDBSCAN* algorithm on N data points has O(N(2)) run-time.”
• Density‐based clustering (2019, methodology): “McInnes and Healy (2017) provide an efficient implementation of HDBSCAN* (Campello et al., 2015), making use of”
• dbscan: Fast Density-Based Clustering with R (2019, methodology): “Since this may result in different clustering results if the data points are processed in a different order, Campello et al. (2015) suggest for DBSCAN* to consider border points as noise.”
• A Matryoshka Hierarchical RAG for Efficient Multi-Hop Question Answering (2026, unknown): “The nodes at the level L represent the documents in D , while each level l L groups the nodes of the level l + 1 into clusters by applying HDBSCAN [1].”