Definition Extraction with LSTM Recurrent Neural Networks

Li, SiLiang; Xu, Bin; Chung, Tong Lee

doi:10.1007/978-3-319-47674-2_16

Cited by 18 publications

(17 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(4) E&S: A system based on more complex dependency-based features (Espinosa-Anke and Saggion, 2014). (5) LSTM-POS: An LSTM-based system which represents each sentence as a mixture of infrequent words and frequent words' associated part-ofspeech (Li et al, 2016).…”

Section: Baselinesmentioning

confidence: 99%

Syntactically Aware Neural Architectures for Definition Extraction

Espinosa-Anke

Schockaert

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

View full text Add to dashboard Cite

Automatically identifying definitional knowledge in text corpora (Definition Extraction or DE) is an important task with direct applications in, among others, Automatic Glossary Generation, Taxonomy Learning, Question Answering and Semantic Search. It is generally cast as a binary classification problem between definitional and non-definitional sentences. In this paper we present a set of neural architectures combining Convolutional and Recurrent Neural Networks, which are further enriched by incorporating linguistic information via syntactic dependencies. Our experimental results in the task of sentence classification, on two benchmarking DE datasets (one generic, one domain-specific), show that these models obtain consistent state of the art results. Furthermore, we demonstrate that models trained on clean Wikipedia-like definitions can successfully be applied to more noisy domain-specific corpora.

show abstract

Section: Baselinesmentioning

confidence: 99%

Syntactically Aware Neural Architectures for Definition Extraction

Espinosa-Anke

Schockaert

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

View full text Add to dashboard Cite

show abstract

“…LSTM-CRF A deep learning model for sequence labeling for DE based on LSTM and CRF (Li, Xu, and Chung 2016).…”

Section: Resultsmentioning

confidence: 99%

“…This table also shows that our model can benefit from the contextualized embeddings (e.g., BERT (Devlin et al 2019)) as BERT can significantly improve the proposed model over three out of four datasets. P R F1 WCL (Navigli and Velardi 2010b) 98.8 60.7 75.2 DefMiner (Jin et al 2013) 92.0 79.0 85.0 B&DC (Boella et al 2014) 88.0 76.0 81.6 E&S (Espinosa- Anke et al 2016) 85.9 85.3 85.4 LSTM-POS (Li, Xu, and Chung 2016) Sequence Classification Performance This section evaluates the models on the sentence classification task for DE. Due to its popularity in the DE literature for this setting, we first report the performance of the models on the general domain dataset WCL.…”

Section: Resultsmentioning

confidence: 99%

“…We use the pre-trained word embeddings GloVe with 300 dimensions from (Pennington, Socher, and Manning 2014) to initialize the model. Finally, to assess how well our model could benefit from the WCL W00 Textbook Contract P R F1 P R F1 P R F1 P R F1 DefMiner (Jin et al 2013) 82.0 78.5 80.5 52.5 49.5 50.5 ------LSTM-CRF (Li, Xu, and Chung 2016) 83.4 80.5 81.7 57.1 55.9 56.2 46.7 47.5 47.0 63.2 73.5 67.4 GCDT (Liu et al 2019) 82.5 81.3 81.2 57.9 56.6 57.4 45.8 48.9 47.0 63.5 73.2 66.7 CVT (Clark et al 2018) 83.0 82.3 82.5 59.3 57.9 58. Table 1: Sentence Labeling Performance.…”

Section: Experiments Dataset and Hyper Parametersmentioning

confidence: 99%

See 1 more Smart Citation

A Joint Model for Definition Extraction with Syntactic Connection and Semantic Consistency

Veyseh

Dernoncourt

Dou

et al. 2020

AAAI

View full text Add to dashboard Cite

Definition Extraction (DE) is one of the well-known topics in Information Extraction that aims to identify terms and their corresponding definitions in unstructured texts. This task can be formalized either as a sentence classification task (i.e., containing term-definition pairs or not) or a sequential labeling task (i.e., identifying the boundaries of the terms and definitions). The previous works for DE have only focused on one of the two approaches, failing to model the inter-dependencies between the two tasks. In this work, we propose a novel model for DE that simultaneously performs the two tasks in a single framework to benefit from their inter-dependencies. Our model features deep learning architectures to exploit the global structures of the input sentences as well as the semantic consistencies between the terms and the definitions, thereby improving the quality of the representation vectors for DE. Besides the joint inference between sentence classification and sequential labeling, the proposed model is fundamentally different from the prior work for DE in that the prior work has only employed the local structures of the input sentences (i.e., word-to-word relations), and not yet considered the semantic consistencies between terms and definitions. In order to implement these novel ideas, our model presents a multi-task learning framework that employs graph convolutional neural networks and predicts the dependency paths between the terms and the definitions. We also seek to enforce the consistency between the representations of the terms and definitions both globally (i.e., increasing semantic consistency between the representations of the entire sentences and the terms/definitions) and locally (i.e., promoting the similarity between the representations of the terms and the definitions). The extensive experiments on three benchmark datasets demonstrate the effectiveness of our approach.1

show abstract

“…DefMiner (Jin et al, 2013) 52.5 / 49.5 / 50.5 ----LSTM-CRF (Li et al, 2016) 57.1 / 55.9 / 56.2 ----GCDT (Liu et al, 2019a) 57.9 / 56.6 / 57. We replace all citations and references to figures, tables, and sections with corresponding placeholders (e.g., CITATION, FIGURE), but keep raw T E X format of mathematical symbols in order to retain the structure of the equations.…”

Section: Error Analysis On Predicted Definitionsmentioning

confidence: 99%

Document-Level Definition Detection in Scholarly Documents: Existing Models, Error Analyses, and Future Directions

Kang¹,

Head²,

Sidhu³

et al. 2020

Proceedings of the First Workshop on Scholarly Document Processing

View full text Add to dashboard Cite

The task of definition detection is important for scholarly papers, because papers often make use of technical terminology that may be unfamiliar to readers. Despite prior work on definition detection, current approaches are far from being accurate enough to use in realworld applications.In this paper, we first perform in-depth error analysis of the current best performing definition detection system and discover major causes of errors. Based on this analysis, we develop a new definition detection system, HEDDEx, that utilizes syntactic features, transformer encoders, and heuristic filters, and evaluate it on a standard sentence-level benchmark. Because current benchmarks evaluate randomly sampled sentences, we propose an alternative evaluation that assesses every sentence within a document. This allows for evaluating recall in addition to precision.HEDDEx outperforms the leading system on both the sentence-level and the document-level tasks, by 12.7 F1 points and 14.4 F1 points, respectively. We note that performance on the high-recall document-level task is much lower than in the standard evaluation approach, due to the necessity of incorporation of document structure as features. We discuss remaining challenges in document-level definition detection, ideas for improvements, and potential issues for the development of reading aid applications.

show abstract

Definition Extraction with LSTM Recurrent Neural Networks

Cited by 18 publications

References 8 publications

Syntactically Aware Neural Architectures for Definition Extraction

Syntactically Aware Neural Architectures for Definition Extraction

A Joint Model for Definition Extraction with Syntactic Connection and Semantic Consistency

Document-Level Definition Detection in Scholarly Documents: Existing Models, Error Analyses, and Future Directions

Contact Info

Product

Resources

About