Extending a CRF-based named entity recognition model for Turkish well formed text and user generated content1

Şeker, Gökhan Akın; Eryiğit, Gülşen

doi:10.3233/sw-170253

Cited by 38 publications

(21 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Consequently, they obtain an F1 score of 48.96% on the noisy dataset. Seker and Eryigit (2017) present the state-of-the-art model on Turkish NER which is another CRF-based model, similar to that of Çelikkaya et al (2013). The authors use an extensive set of morphological and lexical features (e.g., stem, part-of-speech (POS) tags, capitalization, word type and shape flags) and gazetteers.…”

Section: Ner On Turkish Noisy Datamentioning

confidence: 92%

Transfer learning for Turkish named entity recognition on noisy text

Akkaya

Can

2020

Nat. Lang. Eng.

View full text Add to dashboard Cite

In this article, we investigate using deep neural networks with different word representation techniques for named entity recognition (NER) on Turkish noisy text. We argue that valuable latent features for NER can, in fact, be learned without using any hand-crafted features and/or domain-specific resources such as gazetteers and lexicons. In this regard, we utilize character-level, character n-gram-level, morpheme-level, and orthographic character-level word representations. Since noisy data with NER annotation are scarce for Turkish, we introduce a transfer learning model in order to learn infrequent entity types as an extension to the Bi-LSTM-CRF architecture by incorporating an additional conditional random field (CRF) layer that is trained on a larger (but formal) text and a noisy text simultaneously. This allows us to learn from both formal and informal/noisy text, thus improving the performance of our model further for rarely seen entity types. We experimented on Turkish as a morphologically rich language and English as a relatively morphologically poor language. We obtained an entity-level F1 score of 67.39% on Turkish noisy data and 45.30% on English noisy data, which outperforms the current state-of-art models on noisy text. The English scores are lower compared to Turkish scores because of the intense sparsity in the data introduced by the user writing styles. The results prove that using subword information significantly contributes to learning latent features for morphologically rich languages.

show abstract

Section: Ner On Turkish Noisy Datamentioning

confidence: 92%

Transfer learning for Turkish named entity recognition on noisy text

Akkaya

Can

2020

Nat. Lang. Eng.

View full text Add to dashboard Cite

show abstract

“…At present, CRF has been proved to be a better result in the biomedical field, and the machine learning method represented by CRF is called the mainstream. Şeker and Eryiğit presented a CRF‐based NER system, which successfully models the morphologically very rich nature of this language; its extensions to expand the covered named entity types and also to process extra challenging user generated content coming with Web 2.0. The highest accuracy was achieved, ie, 0.82.…”

Section: Related Workmentioning

confidence: 99%

A co‐training based entity recognition approach for cross‐disease clinical documents

Chen

Che

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

Summary Entity recognition plays an important role in building the electronic medical records (EMRs) based medical knowledge graph, which is significant for building Clinical decision support (CDS) system. Cross‐disease clinical documents are context‐related and have different interrelated semantic structures, which bring challenges for entity recognition using traditional methods. In order to solve these problems, this paper proposes a co‐training based entity recognition approach for cross‐disease clinical documents. In this model, we first build partial annotation corpus of the single disease using dependency syntax analysis and the medical statement rule unifies. Then, according to the partial annotation corpus of different diseases, the sentence level features are extracted through the Bi‐LSTM layer with memory unit and CRF methods, which optimize the whole sequence and improve the combination probability of sequence labels. Finally, the results with higher confidence are selected by cross feedback to label the corpus, which enlarges the size of corpus and improves the accuracy of the document entity recognition. The experiment result proves the availability and high efficiency of our method.

show abstract

“…Our first rendition of the MWE annotations also included named entities, but it was decidedly rather primitive. In accordance with more recent studies that aimed to establish annotation standards for MWEs and named entities [11,17,18], we supplied the current releases of IMST and IWT with more comprehensive and systematic MWE annotations, all manually annotated.…”

Section: New Language Resourcesmentioning

confidence: 99%

“…The ITU-METU-Sabancı Treebank (IMST) [9] was later developed as the output of our research project [10], as a reannotated version of the METU-Sabancı Treebank, following a revised annotation framework. IMST proved to be a robust resource [9], despite being a relatively young treebank [11]. Another new resource is the ITU Web Treebank (IWT) [12], which is the first Turkish web treebank and one of the first fully annotated treebanks of user-generated content worldwide, following its international predecessors, the Google English Web Treebank [13] and the French Social Media Bank [14].…”

Section: Introductionmentioning

confidence: 99%

Implementing universal dependency, morphology, and multiword expression annotation standards for Turkish language processing

2018

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Released only a year ago as the outputs of a research project ("Parsing Web 2.0 Sentences", supported in part by a TÜBİTAK 1001 grant (No. 112E276) and a part of the ICT COST Action PARSEME (IC1207)), IMST and IWT are currently the most comprehensive Turkish dependency treebanks in the literature. This article introduces the final states of our treebanks, as well as a newly integrated hierarchical categorization of the multiheaded dependencies and their organization in an exclusive deep dependency layer in the treebanks. It also presents the adaptation of recent studies on standardizing multiword expression and named entity annotation schemes for the Turkish language and integration of benchmark annotations into the dependency layers of our treebanks and the mapping of the treebanks to the latest Universal Dependencies (v2.0) standard, ensuring further compliance with rising universal annotation trends. In addition to significantly boosting the universal recognition of Turkish treebanks, our recent efforts have shown an improvement in their syntactic parsing performance (up to 77.8%/82.8% LAS and 84.0%/87.9% UAS for IMST/IWT, respectively).The final states of the treebanks are expected to be more suited to different natural language processing tasks, such as named entity recognition, multiword expression detection, transfer-based machine translation, semantic parsing, and semantic role labeling.

show abstract

Extending a CRF-based named entity recognition model for Turkish well formed text and user generated content1

Cited by 38 publications

References 32 publications

Transfer learning for Turkish named entity recognition on noisy text

Transfer learning for Turkish named entity recognition on noisy text

A co‐training based entity recognition approach for cross‐disease clinical documents

Implementing universal dependency, morphology, and multiword expression annotation standards for Turkish language processing

Contact Info

Product

Resources

About