Confidence-weighted safe semi-supervised clustering

Explore this paper's citation graph

Summary

Confidence-weighted safe semi-supervised clustering where prior knowledge is given in the form of class labels is proposed where the outputs of the labeled samples with high confidences are restricted to be the given prior labels and those of the local homogeneous unlabeled neighbors modeled by the local graph.

Type
article
Published
2019-05-01
Cited by
36
References
40

Keywords

Cluster analysis, Computer science, Artificial intelligence, Graph, Pattern recognition (psychology)

References

Cited by

Related papers