Neural Open Information Extraction

Cui, Lei; Wei, Furu; Zhou, Ming

doi:10.18653/v1/p18-2065

Cited by 101 publications

(143 citation statements)

References 20 publications

(18 reference statements)

Supporting

Mentioning

128

Contrasting

Order By: Relevance

“…Recently, as a result of their successfulness in a diverse NLP tasks [1], deep neural networks paved the way to the OIE task. A recurrent neural network (RNN) encoder-decoder OIE framework was proposed by Cui et al [20]. A fluctuating length sequence is sent to the network's encoder as a sole input.…”

Section: Neural Network Approachesmentioning

confidence: 99%

Can We Survive without Labelled Data in NLP? Transfer Learning for Open Information Extraction

Sarhan

Spruit

2020

Applied Sciences

View full text Add to dashboard Cite

Various tasks in natural language processing (NLP) suffer from lack of labelled training data, which deep neural networks are hungry for. In this paper, we relied upon features learned to generate relation triples from the open information extraction (OIE) task. First, we studied how transferable these features are from one OIE domain to another, such as from a news domain to a bio-medical domain. Second, we analyzed their transferability to a semantically related NLP task, namely, relation extraction (RE). We thereby contribute to answering the question: can OIE help us achieve adequate NLP performance without labelled data? Our results showed comparable performance when using inductive transfer learning in both experiments by relying on a very small amount of the target data, wherein promising results were achieved. When transferring to the OIE bio-medical domain, we achieved an F-measure of 78.0%, only 1% lower when compared to traditional learning. Additionally, transferring to RE using an inductive approach scored an F-measure of 67.2%, which was 3.8% lower than training and testing on the same task. Hereby, our analysis shows that OIE can act as a reliable source task.

show abstract

Section: Neural Network Approachesmentioning

confidence: 99%

Can We Survive without Labelled Data in NLP? Transfer Learning for Open Information Extraction

Sarhan

Spruit

2020

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Tuple-level matching is used in other Open IE systems Cui et al, 2018). It is done by mapping extracted tuples with their corresponding benchmark tuples.…”

Section: Evaluation Metric and Matching Functionmentioning

confidence: 99%

“…Further, in recent years, there have been efforts to create a supervised Open IE system. (Stanovsky and Dagan, 2016) constructed an annotated corpus for Open IE, and and (Cui et al, 2018) used the annotated data to build a supervised Open IE system by formulating Open IE as sequence tagging and generation problems respectively.…”

Section: Introductionmentioning

confidence: 99%

Supervising Unsupervised Open Information Extraction Models

Roy¹,

Park²,

Lee³

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

We propose a novel supervised open information extraction (Open IE) framework that leverages an ensemble of unsupervised Open IE systems and a small amount of labeled data to improve system performance. It uses the outputs of multiple unsupervised Open IE systems plus a diverse set of lexical and syntactic information such as word embedding, part-ofspeech embedding, syntactic role embedding and dependency structure as its input features and produces a sequence of word labels indicating whether the word belongs to a relation, the arguments of the relation or irrelevant.Comparing with existing supervised Open IE systems, our approach leverages the knowledge in existing unsupervised Open IE systems to overcome the problem of insufficient training data. By employing multiple unsupervised Open IE systems, our system learns to combine the strength and avoid the weakness in each individual Open IE system. We have conducted experiments on multiple labeled benchmark data sets. Our evaluation results have demonstrated the superiority of the proposed method over existing supervised and unsupervised models by a significant margin.

show abstract

“…Our approach is influenced by Wu and Weld (2010), who aligned Wikipedia infobox contents to article text to automatically create training data for an extractor. Recent work on neural extraction models (Cui et al, 2018) has explored entirely supervised models learned from a modified version of the QA-SRL dataset (Stanovsky et al, 2018). The line of research that has most closely examined the prospect of OpenIE-style extractions using webpage structure is the work on Webtables (Cafarella et al, 2008;Dalvi et al, 2012;Balakrishnan et al, 2015;Cafarella et al, 2018).…”

Section: Related Workmentioning

confidence: 99%

Open

Lockard¹,

Shiralkar²,

Dong³

2019

Proceedings of the 2019 Conference of the North

View full text Add to dashboard Cite

Open Information Extraction (OpenIE), the problem of harvesting triples from natural language text whose predicate relations are not aligned to any pre-defined ontology, has been a popular subject of research for the last decade. However, this research has largely ignored the vast quantity of facts available in semistructured webpages. In this paper, we define the problem of OpenIE from semi-structured websites to extract such facts, and present an approach for solving it. We also introduce a labeled evaluation dataset to motivate research in this area. Given a semi-structured website and a set of seed facts for some relations existing on its pages, we employ a semi-supervised label propagation technique to automatically create training data for the relations present on the site. We then use this training data to learn a classifier for relation extraction. Experimental results of this method on our new benchmark dataset obtained a precision of over 70%. A larger scale extraction experiment on 31 websites in the movie vertical resulted in the extraction of over 2 million triples.

show abstract

Neural Open Information Extraction

Cited by 101 publications

References 20 publications

Can We Survive without Labelled Data in NLP? Transfer Learning for Open Information Extraction

Can We Survive without Labelled Data in NLP? Transfer Learning for Open Information Extraction

Supervising Unsupervised Open Information Extraction Models

Open

Contact Info

Product

Resources

About