Towards discipline-independent argumentative zoning: Evidence from chemistry and computational linguistics
Citation: Simone Teufel, Advaith Siddharthan, Colin Batchelor (2009) Towards discipline-independent argumentative zoning: Evidence from chemistry and computational linguistics. Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing (RSS)
Internet Archive Scholar (search for fulltext): Towards discipline-independent argumentative zoning: Evidence from chemistry and computational linguistics
Download: http://portal.acm.org/citation.cfm?id=1699648.1699696
Tagged: Computer Science
(RSS) annotation (RSS), argumentation (RSS), rhetorical structure (RSS), argumentation zoning (RSS)
Summary
This work extends the 'argumentation zoning' described in Simone Teufel's thesis beyond computational linguistics to chemistry, supporting the argument that these zones (being based on rhetorical principles) do not depend on specific domain knowledge. This paper points out that "paradigm shift statements", which are very influential, can be found from rhetorical info in papers.
Along with the argumentative zoning categories (first introduced in An annotation scheme for discourse-level argumentation in research articles) a revised annotation guide for a revised scheme called AZ-II is described; annotation examples are in an appendix. Differences are given; in general AZ-II is more fine-grained. For instance, the OWN work category is divided into methodology, result, and conclusions, as well as recoverable errors. Besides the use in chemistry, the split OWN category has been used in providing feedback to novices who are learning to write abstracts (Feltrim 2005).
The bulk of the paper focuses on the annotation, with particular attention to the role of non-experts. While the aim is discipline-independence, some knowledge of a domain is given, in this case in the form of a 10-page chemistry primer provided to annotators. Further, domain-specific information can determine information such as whether the inability to replicate a published result should be attributed to current authors (OWN_FAIL) or those of the original paper (ANTISUPP).
Related work
See also An annotation scheme for discourse-level argumentation in research articles (which introduces the zones), Discourse-level argumentation in scientific articles: Human and automatic annotation, What's yours and what's mine: Determining intellectual attribution in scientific text, Task-based evaluation of summary quality: Describing relationships between scientific papers, and Argumentative Zoning for improved citation indexing
Argument zoning has also been used to help novices scientists improve their own abstracts (see Argumentative zoning applied to critiquing novices' scientific abstracts).
Selected references
- Christine Chichester, Frdrique Lisacek, Aaron Kaplan, and Agnes Sandor. 2005. Discovering paradigm shift patterns in biomedical abstracts: application to neurodegenerative diseases. In Proceedings of First International Symptosium on Semantic Mining in Biomedicine.
- V. Feltrim, Simone Teufel, Gracas Nunes, and S. Alusio. 2005. Argumentative zoning applied to critiquing novices' scientific abstracts. In Janyce Wiebe, James G. Shanahan, Yan Qu, (eds.) Computing Attitude and Affect in Text: Theory and Applications, pp. 233-245. Springer.
See also
