Active Semi-Supervision for Pairwise Constrained Clustering

By Sugato Basu, Arindam Banerjee, Raymond J. Mooney. SDM, 2004

703 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/95a2ade2834ce6a85d0e8e6b82c81deb34cd115d

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Semi-supervised Kernel Metric Learning Using Relative Comparisons (2016, methodology): “For instance, the ML and CL constraints can be imposed in the form of a Markov random ﬁeld prior over the data items [Basu et al., 2004a,b, Lu and Leen, 2005].”
An Active Learning Approach to Coreference Resolution (2015, methodology): “Here, we note that our approach is inspired from two previously proposed approaches for clustering with pairwise constraints: [Basu et al., 2004a] and [Basu et al., 2004b].”
CECM: Constrained evidential C-means algorithm (2012, background): “They have been incorporated inmany different ways, generally by including a penalty term in the objective function (Basu et al., 2004; Grira et al., 2008) or by altering the distances between objects with respect to the constraints (Xing et al.”
Active co-analysis of a set of shapes (2012, methodology): “Similarly, to minimize user effort, active constrained-clustering has been studied for suggesting pairwise constraints in order to maximize performance [Klein et al. 2002; Basu et al. 2004].”
Semi-supervised graph clustering: a kernel approach (2005, methodology): “However, HMRF-KMEANS can cluster input data only in the form of vectors.”
Clustering with Constraints: Feasibility Issues and the k-Means Algorithm (2005, background): “[5] investigate determining the most informative set of constraints when the algorithm has access to an Oracle.”
Accurate Annotation of Remote Sensing Images via Active Spectral Clustering with Little Expert Knowledge (2015, background): “Most studies on active clustering have built upon traditional clustering methods, such as k-means [36,41,42] and hierarchical clustering [27,37,43].”
An Active Learning Method Using Clustering and Committee-Based Sample Selection for Sound Event Classification (2018, background): “The initialization of medoids is based on farthest-first traversal [21]: a traversed set starts as a singleton of a random segment and the farthest segment to the current traversed set (the distance from a”
A probabilistic framework for semi-supervised clustering (2004, background): “clustering, where the performance of unsupervised clustering algorithms is improved with limited amounts of supervision in the form of labels on the data or constraints [38, 6, 27, 39, 7].”