COVER

About COVER

COVER is a lexical resource providing common-sense knowledge on nouns, and it is automatically generated by combining the lexicographic precision propert to BabelNet and the common-sense hosted in ConceptNet. The reference publication for the resource is [1].

The input for the algorithm generating COVER is obtained via the ClOseSt system.

COVER is a json textual file that contains a series of vectors.

Each vector (exemplar) is organized in five fields, namely:

  • exemplar_label that is the name of the vector (obtained by the input given to the system);
  • exemplar_bsi that is the BabelNet synset ID for the exemplar;
  • exemplar_wnsi that is the WordNet synset ID for the exemplar (set to wn:00000000n if the vector has no WordNet synset ID);
  • exemplar_synsetTerms that is the list of lexicalization for the synset representing the exemplar;
  • exemplar_content that lists the values of the exemplar, organized by dimension.

Each dimension is then organized as follows:

  • dim_name that represents the name of the dimension (the ConceptNet relationship);
  • dim_bsi that represents the BabelNet synset ID of the dimension (it works like an unique identifier for the dimension);
  • dim_values that lists all the values found for that dimension.

Finally, each value (concept) of a dimension is organized in:

  • value_wikititle that represents the Wikipedia Title name for the value concept at hand (set to empty string if the concept has no Wikipedia Title) ;
  • value_bsi that represents the BabelNet synset ID for the concept;
  • value_wnsi that represents the WordNet synset ID for the concept (set to wn:00000000n if the concept has no WordNet synset ID)
  • value_terms that lists the lexicalizations for the concept.

Version 2.1

The version 2.1 relies on the same algorithm adopted in version 2.0, but is built on a much wider input  (31K concepts against the 16K of the version 2.0).

Publication date

27 April 2017

Complete list of relationships
RelatedTo, Synonym, IsA, HasContext, Antonym, FormOf, DerivedFrom, AtLocation, HasA, PartOf, SimilarTo, UsedFor, HasProperty, Causes, CapableOf, HasPrerequisite, HasSubevent, MadeOf, Desires, DBP_Genre, MotivatedByGoal, CreatedBy, Entails, CausesDesire, NotDesires, ReceivesAction, InstanceOf, NotHasProperty, influencedBy, HasLastSubevent, DefinedAs, HasFirstSubevent, DBP_Field, LocatedNear, DBP_KnownFor, NotCapableOf
Statistics about this version

Count
Total of vectors31,837
Total of values605,450
Average dimension population23.97

Download “COVER 2.1”

COVER_v2.1.tar.bz2 – Downloaded 232 times – 41.54 MB

 

Version ABS-COVER

The Version 2.1 is has also been enriched with abstractness values for each value provided by the exemplars. Details on this version can be found in [2].

Publication date

01 May 2018

Download “ABS-COVER 2.1”

COVER_v2.1_ABS.tar.bz2 – Downloaded 240 times –

Version 2.0

The version 2.0 of COVER relies on a hugely improved algorithm.

Specifically:

  • The number of  input concepts has been extended.
  • The new version of ConceptNet (5.5.0) has been adopted.
  • New mechanism for the identification of extracted terms (relies on NASARI Embedded).
  • New threshold of similarity for the extracted terms (set to 0.6).
Publication date

01 April 2017

 

Version 1.1

The version 1.1 of COVER is based on the same algorithm adopted in version 1.0, but is built upon a slightly bigger input.

Detailed information about the resource can be found in [3].

Publication date

02 February 2017

 

Version 1.0

The version 1.0 of COVER relies on a new algorithm. Specifically, the NASARI autoinjection process was added: vectors that would come out empty from the process are now kept in the final resource. They now contain elements from the correspondent NASARI vector, that are injected into the RelatedTo dimension.

Publication date

31 January 2017

 

Version 0.9

The version 0.9 of COVER is the first stable version of the resource.

The resource has been employed in the SemEval 2017 competition task[4].

Publication date

30 January 2017

Creative Commons License All of the above data is licensed under a Creative Commons Attribution 3.0 United States License.

Reference papers

[1] [pdf] [doi] E. Mensa, D. P. Radicioni, and A. Lieto, “COVER: a linguistic resource combining common sense and lexicographic information,” Language Resources and Evaluation, vol. 52, iss. 4, p. 921–948, 2018.
[Bibtex]
@article{mensa2018cover,
author={Mensa, Enrico
and Radicioni, Daniele P.
and Lieto, Antonio},
title={{COVER: a linguistic resource combining common sense and lexicographic information}},
journal={{Language Resources and Evaluation}},
year={2018},
month={Dec},
day={01},
volume={52},
number={4},
pages={921--948},
abstract={Lexical resources are fundamental to tackle many tasks that are central to present and prospective research in Text Mining, Information Retrieval, and connected to Natural Language Processing. In this article we introduce COVER, a novel lexical resource, along with COVERAGE, the algorithm devised to build it. In order to describe concepts, COVER proposes a compact vectorial representation that combines the lexicographic precision characterizing BabelNet and the rich common-sense knowledge featuring ConceptNet. We propose COVER as a reliable and mature resource, that has been employed in as diverse tasks as conceptual categorization, keywords extraction, and conceptual similarity. The experimental assessment is performed on the last task: we report and discuss the obtained results, pointing out future improvements. We conclude that COVER can be directly exploited to build applications, and coupled with existing resources, as well.},
issn={1574-0218},
doi={10.1007/s10579-018-9417-z},
url={https://doi.org/10.1007/s10579-018-9417-z},
pdf={https://rdcu.be/1qFJ}
}
[2] [pdf] “Grasping metaphors: lexical semantics in metaphor analysis,” in The Semantic Web: ESWC 2018 Satellite Events, Cham, 2018, p. 192–195.
[Bibtex]
@inproceedings{mensa18grasping,
editor={Gangemi, Aldo
and Gentile, Anna Lisa
and Nuzzolese, Andrea Giovanni
and Rudolph, Sebastian
and Maleshkova, Maria
and Paulheim, Heiko
and Pan, Jeff Z
and Alam, Mehwish},
title={Grasping Metaphors: Lexical Semantics in Metaphor Analysis},
booktitle= {{The Semantic Web: ESWC 2018 Satellite Events}},
year={2018},
publisher={Springer International Publishing},
address={Cham},
pages={192--195},
abstract={Metaphors represent to date an extraordinary challenge for computational linguistics. Dealing with metaphors has relevant consequences on our ability to build agents and systems that understand Natural Language and text documents: annotating metaphoric constructions by linking the metaphor elements to existing resources is a crucial step to make text documents more easily accessible by machines. Our approach tackles metaphors by considering concepts and their abstractness. We report the encouraging results obtained in a preliminary experimentation; we elaborate on present limitations, and individuate the needed improvements, which will be the base for future work.},
isbn={978-3-319-98192-5},
pdf = {http://delorean.di.unito.it/ls/papers/mensa18grasping.pdf}
}
[3] [pdf] E. Mensa, D. P. Radicioni, and A. Lieto, “TTCS^e: a Vectorial Resource for Computing Conceptual Similarity,” in Proceedings of the 1st Workshop on Sense, Concept and Entity Representations and their Applications, Valencia, Spain, 2017, p. 96–101.
[Bibtex]
@InProceedings{mensa17ttcse,
author = {Mensa, Enrico and Radicioni, Daniele P. and Lieto, Antonio},
title = {{TTCS^e: a Vectorial Resource for Computing Conceptual Similarity}},
booktitle = {{Proceedings of the 1st Workshop on Sense, Concept and Entity Representations and their Applications}},
month = {April},
year = {2017},
address = {Valencia, Spain},
publisher = {Association for Computational Linguistics},
pages = {96--101},
abstract = {In this paper we introduce the TTCS\^{}e, a linguistic resource that relies on
BabelNet, NASARI and ConceptNet, that has now been used to compute the
conceptual similarity between concept pairs. The conceptual representation
herein provides uniform access to concepts based on BabelNet synset IDs, and
consists of a vector-based semantic representation which is compliant with the
Conceptual Spaces, a geometric framework for common-sense knowledge
representation and reasoning. The TTCS\^{}e has been evaluated in a preliminary
experimentation on a conceptual similarity task.},
url = {http://www.aclweb.org/anthology/W17-1912},
pdf = {http://delorean.di.unito.it/ls/papers/mensa17ttcse.pdf}
}
[4] [pdf] E. Mensa, D. P. Radicioni, and A. Lieto, “MERALI at SemEval-2017 Task 2 Subtask 1: a Cognitively Inspired approach,” in Proceedings of the International Workshop on Semantic Evaluation (SemEval 2017), 2017.
[Bibtex]
@InProceedings{mensa17merali,
author = {Mensa, Enrico and Radicioni, Daniele P. and Lieto, Antonio},
title = {{MERALI at SemEval-2017 Task 2 Subtask 1: a Cognitively Inspired approach}},
booktitle={{Proceedings of the International Workshop on Semantic Evaluation (SemEval 2017)}},
year = {2017},
publisher = {{Association for Computational Linguistics}},
abstract = {In this paper we report on the participation of the MeRaLi system to the SemEval Task 2 Subtask 1. The MeRaLi system approaches conceptual similarity through a simple, cognitively inspired, heuristics; it builds on a linguistic resource, the TTCSe, that relies on BabelNet, NASARI and ConceptNet. The linguistic resource in fact contains a novel mixture of common-sense and encyclopedic knowledge. The obtained results point out that there is ample room for improvement, so that they are used to elaborate on present limitations and on future steps.},
pdf = {http://delorean.di.unito.it/ls/papers/mensa17merali.pdf}
}