paper-with-me

Papers

A Relation Extraction Dataset for Knowledge Extraction from Web Tables

2022-10-01 · COLING 2022 10 · Siffi Singh, Alham Fikri Aji, Gaurav Singh, Christos Christodoulopoulos

Relational web-tables are significant sources of structural information that are widely used for relation extraction and population of facts into knowledge graphs. To transform the web-table data into knowledge, we need to identify the relations that exist between column pairs. Currently, there are only a handful of publicly available datasets with relations annotated against natural web-tables. Most datasets are constructed using synthetic tables that lack valuable metadata information, or are limited in size to be considered as a challenging evaluation set. In this paper, we present REDTab, the largest natural-table relation extraction dataset. We have annotated ~9K tables and ~22K column pairs using crowd sourced annotators from MTurk, which has 50x larger number of column pairs than the existing human-annotated benchmark. Our test set is specially designed to be challenging as observed in our experiment results using TaBERT. We publicly release REDTab as a benchmark for the evaluation process in relation extraction.

📄 PDF Abstract BibTeX

Code (1)

alexa/alexa-dataset-redtab 공식 구현

Tasks

Knowledge GraphsRelationRelation Extraction

Similar Papers 제목 키워드 기반

Relation Extraction from Tables using Artificially Generated Metadata

2021-11-16 · ACL ARR November 2021 11 · Anonymous

Relation Extraction (RE) from tables is the task of identifying relations between pairs of columns of a table. Generally, RE models for this task require labelled tables for training. These labelled tables can also be ge…

RelationRelation Extraction

Relation Extraction from Tables using Artificially Generated Metadata

2021-08-24 · Gaurav Singh, Siffi Singh, Joshua Wong, Amir Saffari

Relation Extraction (RE) from tables is the task of identifying relations between pairs of columns of a table. Generally, RE models for this task require labelled tables for training. These labelled tables can also be ge…

RelationRelation Extraction

An Annotated Corpus of Webtables for Information Extraction Tasks

2020-08-18 · Erin Macdonald, Denilson Barbosa

Information Extraction is a well-researched area of Natural Language Processing with applications in web search and question answering concerned with identifying entities and relationships between them as expressed in a …

Question AnsweringRelation ExtractionSentence

TableIE: Capture the Interactions among Joint Information Extraction Explicitly via Double Tables

2021-11-16 · ACL ARR November 2021 11 · Anonymous

Information Extraction (IE) mainly consists of three sub-tasks, Named Entity Recognition, Relation Extraction, and Event Extraction. Although the sub-tasks are highly correlated with each other, most previous works simpl…

Event Extractionnamed-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)+2

Improving Knowledge Base Construction from Robust Infobox Extraction

2019-06-01 · NAACL 2019 6 · Boya Peng, Yejin Huh, Xiao Ling, Michele Banko

A capable, automatic Question Answering (QA) system can provide more complete and accurate answers using a comprehensive knowledge base (KB). One important approach to constructing a comprehensive knowledge base is to ex…

Entity DisambiguationEntity LinkingKnowledge Base ConstructionQuestion Answering+1