README.md

metadata

license: cc-by-nc-2.0
configs:
  - config_name: de
    data_files:
      - split: train
        path: de/argumentative_zoning_multilingual_de_train.jsonl
      - split: test
        path: de/argumentative_zoning_multilingual_de_test.jsonl
  - config_name: en
    data_files:
      - split: train
        path: en/argumentative_zoning_multilingual_en_train.jsonl
      - split: test
        path: en/argumentative_zoning_multilingual_en_test.jsonl
task_categories:
  - text-classification
language:
  - de
  - en

Dataset Card for Argumentative Zoning corpus [AZ corpus]

The dataset originates from the research results of Simone Teufel et al. on the topic of “Argumentative Zoning”.

The English subset was taken from the AZ corpus .az-scixml files and converted to jsonl format. The German subset was further translated using Helsinki-NLP/opus-mt-en-de.

Languages

The supported languages are German and English.

Dataset Structure

• text: full text of the sample
• label: the Text Classification label for the sentence
• label_text: the Text Classification label text for the sentence

Text Classification Tags

Text Classification Tag	Text Classification Label	Description
0	TXT	Indication of paper’s textual structure.
1	CTR	An existing KC is contrasted, compared, or presented as weak.
2	AIM	Statement of research goal.
3	BAS	Existing KC provides basis for new KC.
4	BKG	Description of generally accepted background knowledge.
5	OWN	Description of any other aspect of new KC.
6	OTH	Description of existing KC.

KC = knowledge claim

The text classification labels and their descriptions correspond to the annotation scheme of Teufel et al.

Data Splits

The data is split into two subsets, de for German and en for English. Each subset contains two splits, namely train and test. The split sizes per subset are as follow:

Train	Test
10254	2564

Copyright

All rights reserved by the owners.
The corpus was merely converted into modern data structures and translated into German by the author of this dataset.
The author of this dataset does not claim ownership of this dataset, nor does he distribute it as his own.

The original copyright text reads as follows:

This is the original Argumentative Zoning corpus [AZ corpus], created
and annotated by Simone Teufel and collaborators (Byron
Georgantopolous, Marc Moens, Vasilis Karaiskos, Anne Wilson, Donald
Tennant) between 1996-2004. It consists of 80 AZ-annotated conference
articles in computational linguistics, originally drawn from the Cmplg
arXiv.  The corpus is distributed under the Creative Commons
Attribution-NonCommercial 2.0 UK:England and Wales Licence (CC BY-NC
2.0 UK). 

References

First publication of the AZ corpus was in:

@INPROCEEDINGS{Teufel/Moens:97,
   author          = {Teufel, Simone and Moens, Marc},
   title           = {Sentence extraction as a classification task}, 
   booktitle       = {Proceedings of the ACL/EACL-97 Workshop on Intelligent
                      Scalable Text Summarization},
   year            = 1997,
   editor          = {Mani, Inderjeet and Maybury, Mark T.}
   crossref        = {Mani/Maybury:97},
   bibstate        = {present},
   pages           = {58--65},
}

It is described in detail in the following book [preferred
citation]:

@book{Teufel:10,
  author = 	 {Teufel, Simone},
  title = 	 {The Structure of Scientific Articles: Applications
                  to Citation Indexing and Summarization}, 
  publisher = 	 {CSLI Publications},
  year = 	 2010
}

The argumentative zoning annotation scheme was published in:

@Article{Teufel/Moens:02,
  author = 	 {Teufel, Simone and Marc Moens},
  title = 	 {Summarising Scientific Articles --- Experiments with 
                  Relevance and Rhetorical Status},
  journal = 	  {Computational Linguistics},
  number =        4, 
  volume =        28, 
  pages =         {409--446},
  year = 	  {2002}
}

The SciXML format was first published in:

@InProceedings{Teufel/Elhadad:02,
  author = 	 {Simone Teufel and Noemie Elhadad},
  title = 	 {Collection and linguistic processing of a large-scale 
corpus of medical articles},
  booktitle = 	 {Proceedings of the Third } # LREC # { (LREC 2002)},
  year =	 2002,
  pages =        {1214--1219},
}

Details about the AZ annotation procedure are in the book, first published in the following paper:

@INPROCEEDINGS{Teufel/etal:99,
   author          = {Teufel, Simone and Carletta, Jean and Moens, Marc},
   title           = {An annotation scheme for discourse-level
                      argumentation in research articles},  
   booktitle       = {Proceedings of the Ninth } # EACL # { (EACL-99)}, 
   year            = {1999},
   pages           = {110--117},
}