Dataset Card for CLEF eHealth 2019 Task 1

Dataset Summary

Dataset containing 8,386 non-technical summaries (NTS) of animal experiments recently carried out in Germany (as of September 19, 2018) and originally on-line available at the AnimalTestInfo database. Each NTS contains a title, uses (goals) of the experiments, possible harms caused to the animals, and comments about replacement, reduction and refinement (in the scope of the 3R principles). All documents are in the German language. The dataset includes the ICD-10 codes manually assigned by experts to the NTS. However, some NTSs have no ICD-10 codes assigned to them, as the codes were not applicable to the uses described in the NTS. These are not included in the huggingface dataset. All codes are chapters or groups from the ICD-10 German Modification 2016 version. Finally, the dataset is split into training and development datasets which are meant to be used in the CLEF eHealth 2019, Task 1 - Multilingual Information Extraction.

Supported Tasks and Leaderboards

Mutli-Label Classification.

Languages

German (de)

Dataset Structure

Data Instances

An example of 'train' looks as follows.

{
  "id": "id of NTS",
  "title": "title of NTS",
  "goals": "uses (goals) of the experiments",
  "harms": "possible harms caused to the animals",
  "replacement": "comments about replacement (in the scope of the 3R principles)",
  "reduction": "comments about reduction (in the scope of the 3R principles)",
  "refinement": "comments about refinement (in the scope of the 3R principles),
  "labels": ["J40-J47", "X"]
}

Data Fields

The data fields are the same among all splits.

"id": a string feature.
"title": a string feature.
"goals": a string feature.
"harms": a string feature.
"replacement": a string feature.
"reduction": a string feature.
"refinement": a string feature.
"labels": a list of string features.

Data Splits

The dataset has 3 splits: train, validation, and test.

Dataset Split	Number of Instances in Split
Train	5854
Validation	654
Test	654

Dataset Creation

Curation Rationale

[More Information Needed]

Source Data

Initial Data Collection and Normalization

[More Information Needed]

Who are the source language producers?

[More Information Needed]

Annotations

Annotation process

[More Information Needed]

Who are the annotators?

[More Information Needed]

Personal and Sensitive Information

[More Information Needed]

Considerations for Using the Data

Social Impact of Dataset

[More Information Needed]

Discussion of Biases

[More Information Needed]

Other Known Limitations

[More Information Needed]

Additional Information

Dataset Curators

This data was originally collected by Mariana Neves, Daniel Butzke, Antje Dörendahl, Nora Leich, Barbara Grune, Gilbert Schönfelder of Bundesinstitut für Risikobewertung (BfR).

Licensing Information

The dataset is distributed under the CC BY-NC-ND 4.0 license.

Citation Information

Training and Development/Validation set:

@Misc{openagrar_mods_00046540,
  author = 	{Neves, Mariana
        and Butzke, Daniel
        and D{\"o}rendahl, Antje
        and Leich, Nora
        and Grune, Barbara
        and Sch{\"o}nfelder, Gilbert},
  title = 	{Non-technical Summaries (NTS) of Animal Experiments Indexed with ICD-10 Codes (Version 1.0)},
  year = 	{2019},
  month = 	{Jan},
  day = 	{18},
  publisher = 	{Open Agrar Repository},
  keywords = 	{animal experiment; non-technical summaries; ICD-10 codes; text mining; document indexing; Deutschland},
  abstract = 	{Dataset containing 8,386 non-technical summaries (NTS) of animal experiments recently carried out in Germany (as of September 19, 2018) and originally on-line available at the AnimalTestInfo database (http://animaltestinfo.de).  Each NTS contains a title, uses (goals) of the experiments, possible harms caused to the animals, and comments about replacement, reduction and refinement (in the scope of the 3R principles).  All documents are in the German language. The dataset includes the ICD-10 codes manually assigned by experts to the NTS. However, some NTSs have no ICD-10 codes assigned to them, as the codes were not applicable to the uses described in the NTS. All codes are chapters or groups from the ICD-10 German Modification 2016 version (https://www.dimdi.de/static/de/klassifikationen/icd/icd-10-gm/kode-suche/htmlgm2016/). Finally, the dataset is split into training and development datasets which are meant to be used in the CLEF eHealth 2019, Task 1 - Multilingual Information Extraction (https://sites.google.com/view/clefehealth2019/task-1-multilingual-information-extraction-icd10-coding).},
  doi = 	{10.17590/20190118-134645-0},
  url = 	{https://www.openagrar.de/receive/openagrar_mods_00046540},
  url = 	{https://doi.org/10.17590/20190118-134645-0},
  file = 	{:https://www.openagrar.de/servlets/MCRFileNodeServlet/openagrar_derivate_00019621/nts-icd.zip:TYPE},
  language = 	{en}
}

Test set:

@Misc{openagrar_mods_00049062,
  author = 	{Neves, Mariana
        and Butzke, Daniel
        and D{\"o}rendahl, Antje
        and Leich, Nora
        and Grune, Barbara
        and Sch{\"o}nfelder, Gilbert},
  title = 	{Test set of Non-technical Summaries (NTS) of Animal Experiments Indexed with ICD-10 Codes (Version 1.0)},
  year = 	{2019},
  month = 	{May},
  day = 	{06},
  publisher = 	{Open Agrar Repository},
  keywords = 	{animal experiment; non-technical summaries; ICD-10 codes; text mining; document indexing; Deutschland},
  abstract = 	{This is the official test set of the CLEF eHealth 2019, Task 1 - Multilingual Information Extraction (http://clef-ehealth.org/).
It contains 407 non-technical summaries (NTS) of animal experiments planned to be carried out in Germany and originally online available at the AnimalTestInfo database (http://animaltestinfo.de).  Each NTS contains a title, benefits (goals) of the experiments, possible harms caused to the animals, and comments about replacement, reduction and refinement (in the scope of the 3R principles).  All documents are in the German language.
The dataset includes the ICD-10 codes manually assigned by experts to the NTS. However, some NTSs have no ICD-10 codes assigned to them, as the codes were not applicable to the uses described in the NTS. All codes are chapters or groups from the ICD-10 German Modification 2016 version (https://www.dimdi.de/static/de/klassifikationen/icd/icd-10-gm/kode-suche/htmlgm2016/).},
  doi = 	{10.17590/20190506-101759},
  url = 	{https://www.openagrar.de/receive/openagrar_mods_00049062},
  url = 	{https://doi.org/10.17590/20190506-101759},
  file = 	{:https://www.openagrar.de/servlets/MCRFileNodeServlet/openagrar_derivate_00021578/nts_icd.zip:TYPE},
  language = 	{en}
}

Contributions

Thanks to @heliumind for adding this dataset.