Report on the annotation of semantic roles - TR7 (original) (raw)
Related papers
Putting FrameNet data into the ISO linguistic annotation framework
Proceedings of the ACL 2003 workshop on Linguistic annotation getting the model right -, 2003
This paper describes FrameNet , an online lexical resource for English based on the principles of frame semantics , and considers the FrameNet database in reference to the proposed ISO model for linguistic annotation of language resources (ISO TC37 SC4 ) . We provide a data category specification for frame semantics and FrameNet annotations in an RDF-based language. More specifically, we provide a DAML+OIL markup for lexical units, defined as a relation between a lemma and a semantic frame, and frame-to-frame relations, namely Inheritance and Subframes. The paper includes simple examples of FrameNet annotated sentences in an XML/RDF format that references the project-specific data category specification.
FrameForm: An Open-source Annotation Interface for FrameNet
2021
In this paper, we introduce FrameForm, an open-source annotation tool designed to accommodate predicate annotations based on Frame Semantics. FrameForm is a user-friendly tool for creating, annotating and maintaining computational lexicography projects like FrameNet and has been used while building the Turkish FrameNet. Responsive and open-source, FrameForm can be easily modified to answer the annotation needs of a wide range of different languages.
FrameNet, current collaborations and future goals
Language Resources and Evaluation, 2012
This paper will focus on recent and near-term future developments at FrameNet (FN) and the interoperability issues they raise. We begin by discussing the current state of the Berkeley FN database including major changes in the data format for the latest data release. We then briefly review two recent local projects, ''Rapid Vanguarding'', which has created a new interface for the frame and lexical unit definition process based on the Word Sketch Engine of , and ''Beyond the Core'', which has developed tools for annotating constructions, and created a sample ''construction'' of especially ''interesting'' constructions which are neither simply lexical nor easy for the standard parsers to parse. We also cover two current collaborations, FN's part in the development of the manually annotated subcorpus of the American National Corpus, and a pilot study on aligning WordNet and FrameNet, to exploit the complementary strengths of these quite different resources. We discuss FN-related research on Spanish, Japanese, German (SALSA), Chinese and other languages, and the language-independence of frames, along with interesting FN-related work by others, and a sketch of a large group of imageschematic frames which are now being added to FN. We close with some ideas about how FrameNet can be opened up, to allow broader participation in the development process without losing precision and coherence, including a smallscale study on acquiring data for FN using Amazon's Mechanical Turk crowdsourcing system.
… , mobile Kommunikation und …, 2005
In this paper, we present a rule-based system for the assignment of FrameNet frames by way of a "detour via WordNet". The system can be used to overcome sparse-data problems of statistical systems trained on current FrameNet data. We devise a weighting scheme to select the best frame(s) out of a set of candidate frames, and present first figures of evaluation.
The FrameNet data and software
Proceedings of the 41st Annual Meeting on Association for Computational Linguistics - ACL '03, 2003
The FrameNet project has developed a lexical knowledge base providing a unique level of detail as to the the possible syntactic realizations of the specific semantic roles evoked by each predicator, for roughly 7,000 lexical units, on the basis of annotating more than 100,000 example sentences extracted from corpora. An interim version of the FrameNet data was released in October, 2002 and is being widely used. A new, more portable version of the FrameNet software is also being made available to researchers elsewhere, including the Spanish FrameNet project.
Merging FrameNet and PropBank in a corpus of written Dutch
2000
We discuss the development of a schema for the semantic annotation of a corpus of written Dutch. Our focus is on the annotation of semantic roles. We rely on the proposals made within initiatives such as PropBank and FrameNet. Our aim is to reconcile the PropBank approach to role assignment which is essentially corpus based and syntax driven with the
2003
We refer to these two three-year stages in the life of the project as FrameNet I and FrameNet II.) A one-year extension to cover the period August 2003 to August 2004 has been made possible by financial support from the Defense Advanced Research Projects Agency added to two subcontracts with the National Science Foundation. We gratefully acknowledge the support of Oxford University Press, which, through Timothy Benbow, made it possible for us to use the British National Corpus as the evidential basis for our inquiry into the behavior of English words. Also, through Robert Scriven, OUP gave us permission to select definitions from the Concise English Dictionary to serve as as parts of the FrameNet lexical entries. We are grateful to Bolt Beranek and Newman Corporation for allowing us to use their Identifinder software, which provides high quality named entity recognition as part of our subcorporation process. We are indebted to the Institut für Maschinelle Sprachverarbeitung of the University of Stuttgart, through the kindness of Ulrich Heid and Oliver Christ, for the IMS Corpus Workbench software, which has made it easy for us to extract, explore, and sort example lines and sentences from the BNC. Jerome Feldman was Director of ICSI during most of the FrameNet I period, and is currently head of the Artificial Intelligence group at the Institute; Nelson Morgan has been ICSI Director during the FrameNet II period. Their generosity, along with the efficiency of the technical and administrative staff of this Institute, have made ICSI a welcoming environment for our activities. The Institute's Visitors Program has allowed us, further, to profit from visits between a few months and two years on the part of a number of international researchers-from Spain, Germany, Finland, and Singapore-who participated in the work of the project at various levels, in some cases initiating related efforts elsewhere.
WordNet and FrameNet as complementary resources for annotation
Proceedings of the Third Linguistic Annotation Workshop on - ACL-IJCNLP '09, 2009
WordNet and FrameNet are widely used lexical resources, but they are very different from each other and are often used in completely different ways in NLP. In a case study in which a short passage is annotated in both frameworks, we show how the synsets and definitions of WordNet and the syntagmatic information from FrameNet can complement each other, forming a more complete representation of the lexical semantic of a text than either could alone. Close comparisons between them also suggest ways in which they can be brought into alignment.
FrameNet Meets the Semantic Web: A DAML+OIL Frame Representation
2002
The Berkeley FrameNet Project ) (URL: http:// framenet.icsi.berkeley.edu/˜framenet) is creating an online lexical resource for English, based on the principles of Frame Semantics and supported by corpus evidence. A semantic frame is a script-like structure of inferences, which are linked to the meanings of linguistic units (lexical items). Each frame identifies a set of frame elements (FEs), which are framespecific semantic roles (participants, props, phases of a state of affairs). Our description of each lexical item identifies the frames which underlie a given meaning and the ways in which the FEs are realized in structures headed by the word. The FrameNet database documents the range of semantic and syntactic combinatory possibilities (valences) of each word in each of its senses, through manual annotation of example sentences and automatic summarization of the resulting annotations. The FrameNet database is available in XML, and can be displayed and queried via the web and other interfaces. The FrameNet I data has also been translated into the DAML+OIL extension to XML and the Resource Description Framework (RDF), which can represent our ontologies and to make FrameNet information machine readable and understandable. We have developed an automatic translator from FrameNet 1 data in XML to DAML+OIL; this paper reports on our representation of this data in DAML+OIL.