A Survey of Relation Extraction of Knowledge Graphs

Li, Aoran; Wang, Xinmeng; Wang, Wenhuan; Zhang, Anman; Li, Bohan

doi:10.1007/978-3-030-33982-1_5

Cited by 18 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2017, within the SemEval 2017 Task 10: ScienceIE -Extracting Keyphrases and Relations from Scientific Publications [44] competition, participants were asked to provide tools and methods to find entities and relationships from a scientific annotated corpus of research publications. Since then, approaches for addressing this problem using syntactical patterns and machine learningbased models have been released and employed to transform plain text into structured graph representations [45,46,47]. An example of those is represented by the work described in [47] that was designed to capture the hyponymy relationship between noun phrases that were found in the text.…”

Section: Related Workmentioning

confidence: 99%

SCICERO: A deep learning and NLP approach for generating scientific knowledge graphs in the computer science domain

Dessı̀¹,

Osborne²,

Recupero³

et al. 2022

Knowledge-Based Systems

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

SCICERO: A deep learning and NLP approach for generating scientific knowledge graphs in the computer science domain

Dessı̀¹,

Osborne²,

Recupero³

et al. 2022

Knowledge-Based Systems

View full text Add to dashboard Cite

“…Hence, annotation in RE is the process of chosing relations between head and tail entities in the context from a collection of relation types. Relation Extraction is an indispensable component for composing the knowledge graphs Li et al (2019) that are useful to various downstream NLP applications such as question answering (Dubey, 2021;Saffari et al, 2021;Sen et al, 2021) and dialogue system (Liu et al, 2021b;Chaudhuri et al, 2021;Gao et al, 2021a).…”

Section: Re-annotation In Relation Extractionmentioning

confidence: 99%

Pre-trained Language Models as Re-Annotators

Chen¹

2022

Preprint

View full text Add to dashboard Cite

Annotation noise are widespread in datasets, but manually revising a flawed corpus is time-consuming and error-prone. Hence, given the prior knowledge in Pre-trained Language Models and the expected uniformity across all annotations, we attempt to reduce annotation noise in the corpus through two tasks automatically: (1) Annotation Inconsistency Detection that indicates the credibility of annotations, and (2) Annotation Error Correction that rectifies the abnormal annotations. We investigate how to acquire semantic sensitive annotation representations from Pre-trained Language Models, expecting to embed the examples with identical annotations to the mutually adjacent positions even without fine-tuning. We proposed a novel credibility score to reveal the likelihood of annotation inconsistencies based on the neighbouring consistency. Then, we fine-tune the Pre-trained Language Models based classifier with cross-validation for annotation correction. The annotation corrector is further elaborated with two approaches: (1) soft labelling by Kernel Density Estimation and (2) a novel distant-peer contrastive loss. We study the re-annotation in relation extraction and create a new manually revised dataset, Re-DocRED, for evaluating document-level re-annotation. The proposed credibility scores show promising agreement with human revisions, achieving a Binary F 1 of 93.4 and 72.5 in detecting inconsistencies on TACRED and DocRED respectively.Moreover, the neighbour-aware classifiers based on distant-peer contrastive learning and uncertain labels achieve Macro F 1 up to 66.2 and 57.8 in correcting annotations on TACRED and DocRED respectively. These improvements are not merely theoretical: Rather, automatically denoised training sets demonstrate up to 3.6% performance improvement for state-of-the-art relation extraction models, and the proposed framework is expected to be hundreds of times faster than the human re-annotators empirically.

show abstract

“…Furthermore, within the scholarly domain, extraction of relations from scientific papers has recently raised interest within the SemEval 2017 Task 10: ScienceIE -Extracting Keyphrases and Relations from Scientific Publications [45] and SemEval 2018 Task 7 Semantic Relation Extraction and Classification in Scientific Papers challenge [46], where participants had to face the problem of detecting and classifying domain-specific semantic relations. Since then, extraction methodologies for the purpose to build knowledge graphs from scientific papers started to spread in literature [47,48]. For example, Al-Zaidy et al [49] employed syntactical patterns to detect entities, and defined two types of relations that may exist between two entities (i.e., hyponymy and attributes) by defining rules on noun phrases.…”

Section: Related Workmentioning

confidence: 99%