Domain Adaptation with Structural Correspondence Learning

Discriminative learning methods are widely used in natural language processing. These methods work best when their training and test data are drawn from the same distribution. For many NLP tasks, however, we are confronted with new domains in which labeled data is scarce or non-existent. In such cases, we seek to adapt existing models from a resourcerich source domain to a resource-poor target domain. We introduce structural correspondence learning to automatically induce correspondences among features from different domains. We test our technique on part of speech tagging and show performance gains for varying amounts of source and target training data, as well as improvements in target domain parsing accuracy using our improved tagger.

• Type: Conference
• Year: 2006
• Where: EMNLP
• Authors: John Blitzer, Ryan T. McDonald, Fernando Pereira
• Keywords: Discriminative Learning Methods, EMNLP 2006, EMNLP 2007, Target Domain, Target Training Data