{"title":"Reliability vs. granularity in discourse annotation: What is the trade-off?","authors":"Ludivine Crible, Liesbeth Degand","doi":"10.1515/cllt-2016-0046","DOIUrl":null,"url":null,"abstract":"Abstract We report on the results of an annotation experiment comparing naïve and expert coders in a sense disambiguation task consisting in the assignment of function labels to discourse markers (e.g. well, but, I mean) in spoken French and English using a taxonomy specifically designed for speech. Our qualitative-quantitative assessment of its reliability led us to suggest fundamental revisions of the structure of the taxonomy, striving to find a better balance between reliability and granularity. The resulting model articulates two independent levels of annotation (domains and functions) which, once combined, provide a robust tool for the analysis of discourse markers and relate them to more general functions of spoken language.","PeriodicalId":45605,"journal":{"name":"Corpus Linguistics and Linguistic Theory","volume":"15 1","pages":"71 - 99"},"PeriodicalIF":1.0000,"publicationDate":"2019-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/cllt-2016-0046","citationCount":"30","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Corpus Linguistics and Linguistic Theory","FirstCategoryId":"98","ListUrlMain":"https://doi.org/10.1515/cllt-2016-0046","RegionNum":2,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"0","JCRName":"LANGUAGE & LINGUISTICS","Score":null,"Total":0}
引用次数: 30
Abstract
Abstract We report on the results of an annotation experiment comparing naïve and expert coders in a sense disambiguation task consisting in the assignment of function labels to discourse markers (e.g. well, but, I mean) in spoken French and English using a taxonomy specifically designed for speech. Our qualitative-quantitative assessment of its reliability led us to suggest fundamental revisions of the structure of the taxonomy, striving to find a better balance between reliability and granularity. The resulting model articulates two independent levels of annotation (domains and functions) which, once combined, provide a robust tool for the analysis of discourse markers and relate them to more general functions of spoken language.
期刊介绍:
Corpus Linguistics and Linguistic Theory (CLLT) is a peer-reviewed journal publishing high-quality original corpus-based research focusing on theoretically relevant issues in all core areas of linguistic research, or other recognized topic areas. It provides a forum for researchers from different theoretical backgrounds and different areas of interest that share a commitment to the systematic and exhaustive analysis of naturally occurring language. Contributions from all theoretical frameworks are welcome but they should be addressed at a general audience and thus be explicit about their assumptions and discovery procedures and provide sufficient theoretical background to be accessible to researchers from different frameworks. Topics Corpus Linguistics Quantitative Linguistics Phonology Morphology Semantics Syntax Pragmatics.