Transition-based semantic role labeling with pointer networks

Semantic role labeling (SRL) focuses on recognizing the predicate–argument structure of a sentence and plays critical in many natural language processing tasks such as machine translation question answering. Practically all available methods do not perform full SRL, since they rely pre-identified predicates, most them follow pipeline strategy, using specific models for undertaking one or several SRL subtasks. In addition, previous approaches have strong dependence syntactic information to achieve state-of-the-art performance, despite being trees equally hard produce. These simplifications requirements make majority systems impractical real-world applications. this article, we propose first transition-based approach that is capable completely an input single left-to-right pass, with neither leveraging nor resorting additional modules. Thanks our implementation based Pointer Networks, can be accurately efficiently done O(n2), achieving best performance date languages from CoNLL-2009 shared task.