asahi417's picture
README
dcf770c
|
raw
history blame
25.8 kB
metadata
language:
  - en
license:
  - other
multilinguality:
  - monolingual
size_categories:
  - 1K<n<10K
pretty_name: SemEval2012 task 2 Relational Similarity

Dataset Card for "relbert/semeval2012_relational_similarity_v3"

Dataset Description

Dataset Summary

IMPORTANT: This is the same dataset as relbert/semeval2012_relational_similarity, but with a different dataset construction.

Relational similarity dataset from SemEval2012 task 2, compiled to fine-tune RelBERT model. The dataset contains a list of positive and negative word pair from 89 pre-defined relations. The relation types are constructed on top of following 10 parent relation types.

{
    1: "Class Inclusion",  # Hypernym
    2: "Part-Whole",  # Meronym, Substance Meronym
    3: "Similar",  # Synonym, Co-hypornym
    4: "Contrast",  # Antonym
    5: "Attribute",  # Attribute, Event
    6: "Non Attribute",
    7: "Case Relation",
    8: "Cause-Purpose",
    9: "Space-Time",
    10: "Representation"
}

Each of the parent relation is further grouped into child relation types where the definition can be found here.

Dataset Structure

Data Instances

An example of train looks as follows.

{
  'relation_type': '8d',
  'positives': [ [ "breathe", "live" ], [ "study", "learn" ], [ "speak", "communicate" ], ... ]
  'negatives': [ [ "starving", "hungry" ], [ "clean", "bathe" ], [ "hungry", "starving" ], ... ] 
}

Data Splits

name train validation
semeval2012_relational_similarity 89 89

Number of Positive/Negative Word-pairs in each Split

relation_type level positive (train) negative (train) positive (validation) negative (validation)
1 parent 110 680 129 760
10 parent 60 730 66 823
10a child 10 799 14 894
10a child_prototypical 1 8 1 12
10b child 10 797 13 893
10b child_prototypical 1 8 1 11
10c child 10 800 11 898
10c child_prototypical 1 8 1 9
10d child_prototypical 1 8 1 8
10d child 10 799 10 898
10e child 10 795 8 896
10e child_prototypical 1 8 1 6
10f child 10 799 10 898
10f child_prototypical 1 8 1 8
1a child 10 797 14 892
1a child_prototypical 1 8 1 12
1b child 10 797 14 892
1b child_prototypical 1 8 1 12
1c child_prototypical 1 8 1 9
1c child 10 800 11 898
1d child 10 797 16 890
1d child_prototypical 1 8 1 14
1e child 10 794 8 895
1e child_prototypical 1 8 1 6
2 parent 100 690 117 772
2a child 10 799 15 893
2a child_prototypical 1 8 1 13
2b child_prototypical 1 8 1 9
2b child 10 796 11 894
2c child 10 798 13 894
2c child_prototypical 1 8 1 11
2d child 10 798 10 897
2d child_prototypical 1 8 1 8
2e child 10 799 11 897
2e child_prototypical 1 8 1 9
2f child 10 802 11 900
2f child_prototypical 1 8 1 9
2g child 10 796 16 889
2g child_prototypical 1 8 1 14
2h child_prototypical 1 8 1 9
2h child 10 799 11 897
2i child 10 800 9 900
2i child_prototypical 1 8 1 7
2j child 10 801 10 900
2j child_prototypical 1 8 1 8
3 parent 80 710 80 809
3a child 10 799 11 897
3a child_prototypical 1 8 1 9
3b child 10 802 11 900
3b child_prototypical 1 8 1 9
3c child_prototypical 1 8 1 10
3c child 10 798 12 895
3d child 10 798 14 893
3d child_prototypical 1 8 1 12
3e child 10 802 5 906
3e child_prototypical 1 8 1 3
3f child 10 803 11 901
3f child_prototypical 1 8 1 9
3g child 10 801 6 904
3g child_prototypical 1 8 1 4
3h child_prototypical 1 8 1 8
3h child 10 801 10 900
4 parent 80 710 82 807
4a child 10 802 11 900
4a child_prototypical 1 8 1 9
4b child 10 797 7 899
4b child_prototypical 1 8 1 5
4c child 10 800 12 897
4c child_prototypical 1 8 1 10
4d child_prototypical 1 8 1 2
4d child 10 796 4 901
4e child 10 802 12 899
4e child_prototypical 1 8 1 10
4f child 10 802 9 902
4f child_prototypical 1 8 1 7
4g child 10 798 15 892
4g child_prototypical 1 8 1 13
4h child_prototypical 1 8 1 10
4h child 10 801 12 898
5 parent 90 700 105 784
5a child 10 798 14 893
5a child_prototypical 1 8 1 12
5b child_prototypical 1 8 1 6
5b child 10 801 8 902
5c child 10 799 11 897
5c child_prototypical 1 8 1 9
5d child 10 797 15 891
5d child_prototypical 1 8 1 13
5e child 10 801 8 902
5e child_prototypical 1 8 1 6
5f child 10 801 11 899
5f child_prototypical 1 8 1 9
5g child_prototypical 1 8 1 7
5g child 10 802 9 902
5h child 10 800 15 894
5h child_prototypical 1 8 1 13
5i child 10 800 14 895
5i child_prototypical 1 8 1 12
6 parent 80 710 99 790
6a child 10 798 15 892
6a child_prototypical 1 8 1 13
6b child_prototypical 1 8 1 9
6b child 10 801 11 899
6c child_prototypical 1 8 1 11
6c child 10 801 13 897
6d child 10 804 10 903
6d child_prototypical 1 8 1 8
6e child 10 801 11 899
6e child_prototypical 1 8 1 9
6f child 10 799 12 896
6f child_prototypical 1 8 1 10
6g child 10 798 12 895
6g child_prototypical 1 8 1 10
6h child_prototypical 1 8 1 13
6h child 10 799 15 893
7 parent 80 710 91 798
7a child 10 800 14 895
7a child_prototypical 1 8 1 12
7b child 10 796 7 898
7b child_prototypical 1 8 1 5
7c child 10 797 11 895
7c child_prototypical 1 8 1 9
7d child_prototypical 1 8 1 12
7d child 10 800 14 895
7e child_prototypical 1 8 1 8
7e child 10 797 10 896
7f child 10 796 12 893
7f child_prototypical 1 8 1 10
7g child 10 794 9 894
7g child_prototypical 1 8 1 7
7h child 10 795 14 890
7h child_prototypical 1 8 1 12
8 parent 80 710 90 799
8a child 10 797 14 892
8a child_prototypical 1 8 1 12
8b child_prototypical 1 8 1 5
8b child 10 801 7 903
8c child 10 796 12 893
8c child_prototypical 1 8 1 10
8d child 10 796 13 892
8d child_prototypical 1 8 1 11
8e child 10 796 11 894
8e child_prototypical 1 8 1 9
8f child 10 797 12 894
8f child_prototypical 1 8 1 10
8g child_prototypical 1 8 1 5
8g child 10 793 7 895
8h child 10 798 14 893
8h child_prototypical 1 8 1 12
9 parent 90 700 96 793
9a child 10 795 14 890
9a child_prototypical 1 8 1 12
9b child 10 799 12 896
9b child_prototypical 1 8 1 10
9c child 10 790 7 892
9c child_prototypical 1 8 1 5
9d child_prototypical 1 8 1 7
9d child 10 803 9 903
9e child 10 804 8 905
9e child_prototypical 1 8 1 6
9f child 10 799 10 898
9f child_prototypical 1 8 1 8
9g child 10 796 14 891
9g child_prototypical 1 8 1 12
9h child 10 799 13 895
9h child_prototypical 1 8 1 11
9i child 10 799 9 899
9i child_prototypical 1 8 1 7

Citation Information

@inproceedings{jurgens-etal-2012-semeval,
    title = "{S}em{E}val-2012 Task 2: Measuring Degrees of Relational Similarity",
    author = "Jurgens, David  and
      Mohammad, Saif  and
      Turney, Peter  and
      Holyoak, Keith",
    booktitle = "*{SEM} 2012: The First Joint Conference on Lexical and Computational Semantics {--} Volume 1: Proceedings of the main conference and the shared task, and Volume 2: Proceedings of the Sixth International Workshop on Semantic Evaluation ({S}em{E}val 2012)",
    month = "7-8 " # jun,
    year = "2012",
    address = "Montr{\'e}al, Canada",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/S12-1047",
    pages = "356--364",
}