An experimental study of constrained clustering effectiveness in presence of erroneous constraints

Ares, M. Eduardo; Parapar, Javier; Barreiro, Álvaro

doi:10.1016/j.ipm.2011.08.006

Cited by 3 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such human judgment errors are especially likely when multiple experts are used to arrive at a consensus judgment. As the accuracy of constraints imposed to the algorithm ultimately impacts clustering accuracy (Ares et al, 2012), and that inaccuracy of constraints can occur due to human judgment errors and is an important problem, methods that can help users identify which constraints are likely to be subject to errors should be helpful in improving accuracy (Anil et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

A Lagrangian-based score for assessing the quality of pairwise constraints in semi-supervised clustering

Randel

Aloise

Blanchard

et al. 2021

Data Min Knowl Disc

View full text Add to dashboard Cite

Clustering algorithms help identify homogeneous subgroups from data. In some cases, additional information about the relationship among some subsets of the data exists. When using a semi-supervised clustering algorithm, an expert may provide additional information to constrain the solution based on that knowledge and, in doing so, guide the algorithm to a more useful and meaningful solution. Such additional information often takes the form of a cannot-link constraint (i.e., two data points cannot be part of the same cluster) or a must-link constraint (i.e., two data points must be part of the same cluster). A key challenge for users of such constraints in semi-supervised learning algorithms, however, is that the addition of inaccurate or conflicting constraints can decrease accuracy and little is known about how to detect whether expert-imposed constraints are likely incorrect. In the present work, we introduce a method to score each must-link and cannot-link pairwise constraint as likely incorrect. Using synthetic experimental examples and real data, we show that the resulting impact score can successfully identify individual constraints that should be removed or revised.

show abstract

Section: Introductionmentioning

confidence: 99%