Paper: Online Methods for Multi-Domain Learning and Adaptation

ACL ID D08-1072
Title Online Methods for Multi-Domain Learning and Adaptation
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2008
Authors

NLP tasks are often domain specific, yet sys- tems can learn behaviors across multiple do- mains. We develop a new multi-domain online learning framework based on parameter com- bination from multiple classifiers. Our algo- rithms draw from multi-task learning and do- main adaptation to adapt multiple source do- main classifiers to a new target domain, learn across multiple similar domains, and learn across a large number of disparate domains. We evaluate our algorithms on two popular NLP domain adaptation tasks: sentiment clas- sification and spam filtering.