Improving LSTM CRFs using character-based compositions for Korean named entity recognition. (March 2019)

• Record Type: Journal Article
• Authors: Na, Seung-Hoon; Kim, Hyun; Min, Jinwoo; Kim, Kangil
• Is Part Of: Computer speech & language. Volume 54(2019)
• Page Start: 106
• Page End: 121
• Publication Date: 2019-03
• Subjects: Named entity recognition -- Long short term memory -- Convolutional neural networks -- Character-based composition; Speech processing systems -- Periodicals; Automatic speech recognition -- Periodicals; Computers -- Periodicals; Linguistics -- Periodicals; Speech-Language Pathology -- Periodicals; Traitement automatique de la parole -- Périodiques; Reconnaissance automatique de la parole -- Périodiques; Automatic speech recognition; Speech processing systems; Electronic journals; Periodicals; 006.454
• DOI: 10.1016/j.csl.2018.09.005
• Languages: English
• ISSNs: 0885-2308

Abstract

Highlights: We explore the use of character-based LSTM CRF for Korean NER task. We propose a hybrid character-based word representation using ConvNet and LSTM. The proposed hybrid representation improves each single word presentation. Abstract: Standard approaches to named entity recognition (NER) are based on sequential labeling methods, such as conditional random fields (CRFs), which label each word in a sentence and extract entities from them that correspond to named entities. With the extensive deployment of deep learning methods for sequential labeling tasks, state-of-the-art NER performance has been achieved on long short-term memory (LSTM) architectures using only basic features. In this paper, we address Korean NER tasks and propose an extension of a bidirectional LSTM CRF by investigating character-based representation. Our extension involves deploying a hybrid representation using ConvNet and LSTM for the sequential modeling of characters, namely a character-based LSTM-ConvNet hybrid representation . Using morphemes as processing units for bidirectional LSTM, we apply a proposed hybrid representation composed of morpheme vectors. Experimental results showed that the proposed LSTM-ConvNet hybrid representation yielded improvements over each single representation on standard Korean NER tasks.