Dependency Parsing Domain Adaptation using Transductive SVM

Dependency Parsing domain adaptation involves adapting a dependency parser, trained on an annotated corpus from a given domain (e.g., newspaper articles), to work on a different target domain (e.g., legal documents), given only an unannotated corpus from the target domain. We present a shift/reduce dependency parser that can handle unlabeled sentences in its training set using a transductive SVM as its action selection classifier. We illustrate the the experiments we performed with this parser on a domain adaptation task for the Italian language.