Proceedings of the 2017 ACM on Conference on Information and Knowledge Management 2017
DOI: 10.1145/3132847.3133105
|View full text |Cite
|
Sign up to set email alerts
|

A Robust Named-Entity Recognition System Using Syllable Bigram Embedding with Eojeol Prefix Information

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
4
0

Year Published

2018
2018
2022
2022

Publication Types

Select...
3
1

Relationship

0
4

Authors

Journals

citations
Cited by 4 publications
(4 citation statements)
references
References 10 publications
0
4
0
Order By: Relevance
“…Although the corpus does not aim at medical NER, we expect that this extra experiment would help to justify the effectiveness of BERT for Korean NER. We choose bi-LSTM-CRF as a benchmark model because the model has achieved state-of-the-art performance in Korean NER [20,22]. The architecture of bi-LSTM-CRF is shown in Fig.…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…Although the corpus does not aim at medical NER, we expect that this extra experiment would help to justify the effectiveness of BERT for Korean NER. We choose bi-LSTM-CRF as a benchmark model because the model has achieved state-of-the-art performance in Korean NER [20,22]. The architecture of bi-LSTM-CRF is shown in Fig.…”
Section: Resultsmentioning
confidence: 99%
“…However, because of the linguistic property that the words are not always clearly separated, the tokenization and input encoding influence a lot to the final performance in general. The character-level n-gram encoding with additional linguistic information is one of the state-of-the-art approaches for Korean NER [ 20 ]. A recent work reports that jamo (Korean alphabet) level representation extracts well the word semantics in terms of word similarity [ 21 ].…”
Section: Introductionmentioning
confidence: 99%
“…Lee et al (2020) explored both the syllable level and sub-character level representations of the text, achieving similar results to multilingual BERT with 1/10 of the training data. Kwon et al (2017) proposed a deep learning based NER system that operates over syllables rather than words, resulting in a speedup by removing the need for morphological analysis. Kim et al (2021) resentations and also found that syllables were the most effective representation for Korean NER.…”
Section: Related Workmentioning
confidence: 99%
“…Recently, ML-based systems that implement well-known supervised learning models have been developed to improve the accuracy of NER systems. These models include: Decision Trees (DT) [4], Maximum Entropy Models (MEM) [5], Conditional Random Fields (CRF) [6,7], structural Support Vector Machines (SVM) [8], and recent neural network models based on Long-Short Term Memory (LSTM) with a CRF layer [11][12][13].…”
Section: Previous Workmentioning
confidence: 99%