D'Souza, Jennifer
Auer, Soeren
Pedersen, Ted
2021-05-04
<p><a href="https://ncg-task.github.io/"><strong>NLPContributionGraph</strong></a> was introduced as Task 11 at <a href="https://semeval.github.io/SemEval2021/tasks">SemEval 2021</a> for the first time. The task is defined on a dataset of Natural Language Processing (NLP) scholarly articles with their contributions structured to be integrable within Knowledge Graph infrastructures such as the <a href="https://www.orkg.org/">Open Research Knowledge Graph</a>. The structured contribution annotations are provided as (1) <strong>Contribution sentences</strong> : a set of sentences about the contribution in the article; (2) <strong>Scientific terms and relations</strong>: a set of scientific terms and relational cue phrases extracted from the contribution sentences; and (3) <strong>Triples</strong>: semantic statements that pair scientific terms with a relation, modeled toward subject-predicate-object RDF statements for KG building. The Triples are organized under three (mandatory) or more of twelve total information units (viz., <em>ResearchProblem</em>, <em>Approach</em>, <em>Model</em>, <em>Code</em>, <em>Dataset</em>, <em>ExperimentalSetup</em>, <em>Hyperparameters</em>, <em>Baselines</em>, <em>Results</em>, <em>Tasks</em>, <em>Experiments</em>, and <em>AblationAnalysis</em>).</p>
<p><strong>The Shared Task</strong></p>
<p>As a complete submission for the Shared Task, given NLP scholarly articles in plaintext format, systems had to automatically extract the following information: <em> contribution sentences; </em> scientific term and predicate phrases from the sentences; and (subject,predicate,object) triple statements toward KG building organized under three or more of twelve total information units. The shared task has an open evaluation never-ending official online evaluation at <a href="https://competitions.codalab.org/competitions/25680">Codalab</a>.</p>
https://doi.org/10.25835/0022787
oai:zenodo.org:4737071
eng
Zenodo
https://zenodo.org/communities/semeval
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
SemEval, International Workshop on Semantic Evaluation, Bangkok (online), 5-6 August, 2021
SemEval-2021
language resource
English
natural language processing
NLP
semantics corpus
shared task resource
open research knowledge graph
scholarly knowledge graphs
open science graphs
knowledge representation
scholarly articles
SemEval-2021 Task 11: NLPContributionGraph
info:eu-repo/semantics/other