CoBoosting

CoBoost is a variant of Boosting proposed by Collins and Singer.^[1]

It may be seen as a combination of co-training and boosting. Each example is available in two views, and boosting is applied iteratively and alternatively in each views using pseudo-labels produced in the other view.

References

Footnotes

1. ^ Michael Collins and Yoram Singer, Unsupervised Models for Named Entity Classification. Proceedings of the 1999 Joint SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora, pp. 100-110, 1999.