tailieunhanh - Báo cáo khoa học: "A FrameNet-based Semantic Role Labeler for Swedish"

We present a FrameNet-based semantic role labeling system for Swedish text. As training data for the system, we used an annotated corpus that we produced by transferring FrameNet annotation from the English side to the Swedish side in a parallel corpus. In addition, we describe two frame element bracketing algorithms that are suitable when no robust constituent parsers are available. We evaluated the system on a part of the FrameNet example corpus that we translated manually, and obtained an accuracy score of on the classification of presegmented frame elements, and precision and recall scores of and . | A FrameNet-based Semantic Role Labeler for Swedish Richard Johansson and Pierre Nugues Department of Computer Science LTH Lund University Sweden richard pierre @ Abstract We present a FrameNet-based semantic role labeling system for Swedish text. As training data for the system we used an annotated corpus that we produced by transferring FrameNet annotation from the English side to the Swedish side in a parallel corpus. In addition we describe two frame element bracketing algorithms that are suitable when no robust constituent parsers are available. We evaluated the system on a part of the FrameNet example corpus that we translated manually and obtained an accuracy score of on the classification of presegmented frame elements and precision and recall scores of and for the complete task. 1 Introduction Semantic role labeling SRL the process of automatically identifying arguments of a predicate in a sentence and assigning them semantic roles has received much attention during the recent years. SRL systems have been used in a number of projects in Information Extraction and Question Answering and are believed to be applicable in other domains as well. Building SRL systems for English has been studied widely Gildea and Jurafsky 2002 Litkowski 2004 inter alia. However all these works rely on corpora that have been produced at the cost of a large effort by human annotators. For instance the current FrameNet corpus Baker et al. 1998 consists of 130 000 manually annotated sentences. For smaller languages such as Swedish such corpora are not available. In this work we describe a FrameNet-based semantic role labeler for Swedish text. Since there was no existing training corpus available no FrameNet-annotated Swedish corpus of substantial size exists we used an English-Swedish parallel corpus whose English part was annotated with semantic roles using the FrameNet annotation scheme. We then applied a cross-language transfer to derive an annotated .