Constrained K-means Clustering with Background Knowledge

By Kiri L. Wagstaff, Claire Cardie, Seth Rogers, Stefan Schrödl. International Conference on Machine Learning, 2001

2,961 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/0bacca0993a3f51649a6bb8dbb093fc8d8481ad4

How researchers see this paper

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

Citations in the corpus

Hierarchical Clustering of a Mixture Model (2004, methodology): “Several algorithms that perform clustering of data points given such constraints were recently proposed [ 11 , 5, 12].”
Sigma-point kalman filters for probabilistic inference in dynamic state-space models (2004, background): “In order to robustly address this problem one can make use of more advanced constrained clustering algorithms [198, 103], which, for example, makes use of convexity constraints and other meta-information derived from the environment map, to constrain the clustering process.”
Multimodal Motion Prediction with Stacked Transformers (2021, methodology): “The ﬁrst one adopts constrained K-means [32], while we manually split the space into fan-shaped regions(similar with Fig 3) for the another one.”
Adaptive computation and machine learning (1998, unknown): “Constraint-based methods use the constraint-based and distance-based methods provided supervision to guide the algorithm toward a data partitioning that avoids violating the constraints (Demiriz et al., 1999; Wagstaff et al., 2001; Basu et al., 2002).”
Cryptocurrency trading: a comprehensive survey (2020, methodology): “K-means stores the k-centroids used to define the clusters; a point is considered to be in a particular cluster if it is closer to the cluster’s centroid than any other centroid [245].”
From Instance-level Constraints to Space-Level Constraints: Making the Most of Prior Knowledge in Data Clustering (2002, methodology): “In what follows, CCL is the constrained completelink algorithm presented above and CKM is our reimplementation of COP- -means (Wagstaff et al., 2001).”
Active Semi-Supervision for Pairwise Constrained Clustering (2004, methodology): “In the initialization step of PCKMeans, we take the transitive closure of the must-link constraints [33] and augment the set by adding these entailed constraints.”
Smart Tourist - Passive Mobility Tracking Through Mobile Application (2014, methodology): “Clustering data has been focused in many applications including data mining, statistical data analysis, computation and vector quantisation [9], pattern recognition [10] etc.”
Exploratory Learning (2013, background): “There is also a substantial body of work on constrained clustering; for instance, Wagstaff et al [26] describe a constrained clustering variant of K-Means “must-link” and “cannot-link” constraints between pairs.”
UAV-YOLO: Small Object Detection on Unmanned Aerial Vehicle Perspective (2020, methodology): “During training, k-means [36,37] is used to cluster different numbers of anchor boxes to find the optimized number and size of them.”
Constrained 1-Spectral Clustering (2012, unknown): “ven similarities alone or there is more than one way to group the given items. In either case, clustering methods benet from domain knowledge that gives bias to the desired clustering. Wagsta et. al [17] are the rst to consider constrained clustering by encoding available domain knowledge in the for”
Centroid Networks for Few-Shot Clustering and Unsupervised Few-Shot Classification (2019, methodology): “Some methods take the pairwise supervision as hard constraints (Wagstaff et al., 2001), while others (including CCN) learn metrics which tend to satisfy those constraints (Bilenko et al., 2004).”
Correlation clustering in general weighted graphs (2006, background): “2 As a Way of Solving Constraint-Clustering Problems There has been a great deal of work on solving a clustering problem subject to constraints [18, 29, 30, 15].”