Filters








926 Hits in 3.4 sec

LM-Based Word Embeddings Improve Biomedical Named Entity Recognition: A Detailed Analysis [chapter]

Liliya Akhtyamova, John Cardiff
2020 Lecture Notes in Computer Science  
We derive the contextualized word embeddings from the Flair framework and apply them to the task of biomedical NER on 5 benchmark datasets, yielding major improvements over the baseline and achieving competitive  ...  Recent studies have shown that contextualized word embeddings outperform other types of embeddings on a variety of tasks.  ...  These advantages of contextualized word embeddings motivate us to apply them to biomedical NER tasks.  ... 
doi:10.1007/978-3-030-45385-5_56 fatcat:mshccwr5wrbxllfsnnvlaw4vqm

Testing Contextualized Word Embeddings to Improve NER in Spanish Clinical Case Narratives

Liliya Akhtyamova, Paloma Martinez, Karin Verspoor, John Cardiff
2020 IEEE Access  
In the Big Data era, there is an increasing need to fully exploit and analyze the huge quantity of information available about health.  ...  In this work, we develop a deep learning-based NLP pipeline for biomedical entity extraction in Spanish clinical narratives.  ...  Her research focuses on biomedical text mining and clinical data analysis.  ... 
doi:10.1109/access.2020.3018688 fatcat:u5veccj2ovebdh32obx3mcegtu

Deep Neural Model with Contextualized-word Embeddings for Named Entity Recognition in Spanish Clinical Text

Renzo M. Rivera Zavala, Paloma Martínez
2020 Annual Conference of the Spanish Society for Natural Language Processing  
The approach was evaluated on the CANTEMIST Corpus obtaining an F-measure of 82.3% for NER.  ...  We propose a deep neural approach based on two Bidirectional Long Short-Term Memory (Bi-LSTM) network and Conditional Random Field (CRF) network using character and contextualized-word embeddings to deal  ...  The motivation would be to see whether contextualized-word representations generated with biomedical knowledge can help to improve the results and provide a deep learning model for biomedical NER and concept  ... 
dblp:conf/sepln/ZavalaM20 fatcat:5q3l3r4kv5gutanqv5xhtip3me

Probing Biomedical Embeddings from Language Models [article]

Qiao Jin, Bhuwan Dhingra, William W. Cohen, Xinghua Lu
2019 arXiv   pre-print
Surprisingly, while fine-tuned BioBERT is better than BioELMo in biomedical NER and NLI tasks, as a fixed feature extractor BioELMo outperforms BioBERT in our probing tasks.  ...  Contextualized word embeddings derived from pre-trained language models (LMs) show significant improvements on downstream NLP tasks.  ...  However, unlabeled biomedical texts are abundant, and their full potential has perhaps not yet been fully realized.  ... 
arXiv:1904.02181v1 fatcat:slqbi5ak4nbo5c4st4adnd4ul4

Hero-Gang Neural Model For Named Entity Recognition [article]

Jinpeng Hu, Yaling Shen, Yang Liu, Xiang Wan, Tsung-Hui Chang
2022 arXiv   pre-print
Recently, since the multi-head attention mechanism applied in the Transformer model can effectively capture longer contextual information, Transformer-based models have become the mainstream methods and  ...  Named entity recognition (NER) is a fundamental and important task in NLP, aiming at identifying named entities (NEs) from free text.  ...  This is because multiple-level features can be reasonably encoded Biomedical NER We also compare our model with state-of-the-art models in the biomedical NER on the aforementioned datasets with all results  ... 
arXiv:2205.07177v1 fatcat:avkbol4a6rbbxk4af5jd4smjyy

Analyzing transfer learning impact in biomedical cross-lingual named entity recognition and normalization

Renzo M. Rivera-Zavala, Paloma Martínez
2021 BMC Bioinformatics  
Recent NER approaches use contextualized word representations as input for a downstream classification task.  ...  Moreover, contextualized representations help to understand complexities and ambiguity inherent to biomedical texts.  ...  The motivation would be to see whether contextualized word representations generated with biomedical knowledge can help to improve the results and provide a deep learning model for biomedical NER and concept  ... 
doi:10.1186/s12859-021-04247-9 pmid:34920703 pmcid:PMC8680060 fatcat:dsavlikunvc4zeejhsl74hteo4

Biomedical and Clinical English Model Packages in the Stanza Python NLP Library [article]

Yuhao Zhang, Yuhui Zhang, Peng Qi, Christopher D. Manning, Curtis P. Langlotz
2020 arXiv   pre-print
These packages offer accurate syntactic analysis and named entity recognition capabilities for biomedical and clinical text, by combining Stanza's fully neural architecture with a wide variety of open  ...  datasets as well as large-scale unsupervised biomedical and clinical text data.  ...  Its fully neural pipeline design allows us to extend its language processing capabilities to the biomedical and clinical domain.  ... 
arXiv:2007.14640v1 fatcat:i2opcouxzjgrdehxjibw2iuzmq

TaxoNERD: deep neural models for the recognition of taxonomic entities in the ecological and evolutionary literature [article]

Nicolas Le Guillarme, Wilfried Thuiller
2021 bioRxiv   pre-print
In recent years, deep learning-based NER systems have become ubiqutous, yielding state-of-the-art results in the general and biomedical domains.  ...  To address this issue, we leverage existing DNN models pretrained on large biomedical corpora using transfer learning.  ...  For all these reasons, we choose to build upon the spaCy library to create our taxonomic NER Non-contextual embeddings.  ... 
doi:10.1101/2021.06.08.444426 fatcat:l2epu7suznc4taiknapvf73j44

Exploiting the contextual cues for bio-entity name recognition in biomedical literature

Zhihao Yang, Hongfei Lin, Yanpeng Li
2008 Journal of Biomedical Informatics  
To extract biomedical information about bio-entities from the huge amount of biomedical literature, the first key step is recognizing their names in these literatures, which remains a challenging task  ...  Conditional Random Field-based approach used to recognize the names of bio-entities including gene, protein, cell type, cell line and studies the methods of improving the performance by the exploitation of the contextual  ...  These show the performance of NER in the biomedical domain is far below the one of NER in the general domain.  ... 
doi:10.1016/j.jbi.2008.01.002 pmid:18272430 fatcat:4ge66aqjpzfs5b37lbryfv2uua

Contextualized French Language Models for Biomedical Named Entity Recognition

Jenny Copara, Julien Knafou, Nona Naderi, Claudia Moro, Patrick Ruch, Douglas Teodoro
2020 Traitement Automatique des Langues Naturelles & Rencontre des Étudiants Chercheurs en Informatique pour le Traitement Automatique des Langues  
Modèles contextualisés en langue française pour la reconnaissance des entités nommées dans le domaine biomédical La reconnaissance des entités nommées (NER) est essentielle pour les applications biomédicales  ...  Dans ce travail, nous explorons les modèles de langage contextualisés pour la NER dans les textes biomédicaux français dans le cadre du Défi Fouille de Textes.  ...  Each NER model is a BERT module with a fully connected layer on top of the hidden states of each token.  ... 
dblp:conf/taln/CoparaKNMRT20 fatcat:e2mex7el2fdllbukksvu7wnmtq

Integrating Dictionary Feature into A Deep Learning Model for Disease Named Entity Recognition [article]

Hamada A. Nayel, Shashrekha H. L
2019 arXiv   pre-print
Word embeddings trained over general domain texts as well as biomedical texts have been used to represent input to the proposed model.  ...  This study also compares two different Segment Representation (SR) schemes, namely IOB2 and IOBES for Disease-NER.  ...  A fully connected neural network has been used to convert the contextual representations of the tokens to a vector where each entry corresponds to a score for each output tag.  ... 
arXiv:1911.01600v1 fatcat:ob26ush4xvfq5pw5rmcopfd25e

Domain specific BERT representation for Named Entity Recognition of lab protocol [article]

Tejas Vaidhya, Ayush Kaushal
2020 arXiv   pre-print
For instance, the BERT family seems to work exceptionally well on the downstream task from NER tagging to the range of other linguistic tasks.  ...  contains a lot of different tokens used only in the medical industry such as the name of different diseases, devices, organisms, medicines, etc. that makes it difficult for traditional BERT model to create contextualized  ...  ., 2017) have greatly improved performance in biomedical named entity recognition (NER) over the last few years.  ... 
arXiv:2012.11145v1 fatcat:brekslazmbdnnhnsb2nzkmq2xm

FlexNER: A Flexible LSTM-CNN Stack Framework for Named Entity Recognition [article]

Hongyin Zhu, Wenpeng Hu, Yi Zeng
2019 arXiv   pre-print
Named entity recognition (NER) is a foundational technology for information extraction. This paper presents a flexible NER framework compatible with different languages and domains.  ...  We conduct experiments on five languages, such as English, German, Spanish, Dutch and Chinese, and biomedical fields, such as identifying the chemicals and gene/protein terms from scientific works.  ...  Results on Biomedical NER In biomedical domain, one of the challenges is the limited size of training data.  ... 
arXiv:1908.05009v1 fatcat:zgl5pf5pyvghxjyo3oru7pukgq

Biomedical and clinical English model packages for the Stanza Python NLP library

Yuhao Zhang, Yuhui Zhang, Peng Qi, Christopher D Manning, Curtis P Langlotz
2021 JAMIA Journal of the American Medical Informatics Association  
The resulting pipelines are fully based on neural networks, and are able to perform tokenization, part-of-speech tagging, lemmatization, dependency parsing, and named entity recognition for both biomedical  ...  For NER, our systems substantially outperform scispaCy, and are better or on par with the state-of-the-art performance from BioBERT, while being much more computationally efficient.  ...  NER models Stanza's NER component adopts the architecture of the contextualized string representation-based sequence tagger. 27 For each domain, we train a forward and a backward LSTM character-level  ... 
doi:10.1093/jamia/ocab090 pmid:34157094 fatcat:6b3ncqkyxvezxk2lnv6xicy5pe

Publicly Available Clinical BERT Embeddings [article]

Emily Alsentzer, John R. Murphy, Willie Boag, Wei-Hung Weng, Di Jin, Tristan Naumann, Matthew B. A. McDermott
2019 arXiv   pre-print
Contextual word embedding models such as ELMo (Peters et al., 2018) and BERT (Devlin et al., 2018) have dramatically improved performance for many natural language processing (NLP) tasks in recent months  ...  They find the specificity offered by biomedical texts translated to improved performance on several biomedical NLP tasks, and fully release their pre-trained BERT model.  ...  Contextual Clinical & Biomedical Embeddings Several works have explored the utility of contextual models in the clinical and biomedical domains.  ... 
arXiv:1904.03323v3 fatcat:2sbf755lgresfiq7hosmw6nd2e
« Previous Showing results 1 — 15 out of 926 results