안재현
소속 KT 융합기술원 연구원(한국어 처리 연구)
경력 <학술지>
* Hokyung Lee, Jaehyun An, Jungmin Yoon, Kyoungman Bae and
Youngjoong Ko. A Method of Solving Entity Linking Ambiguity and NIL
Entity Recognition for efficient Entity Linking based on Wikipedia.
Journal of Korean Institute of Information Scientists and Engineers
(KIISE), Vol. 44, No. 8, pp. 813- 821, August 2017. (ISSN 2383- 630X)
*Jaehyun An and Youngjoong Ko. Korean Dependency Parser Using
Dependency Relation Tag Distribution. Journal of KIISE Transactions on
Computing Practices, Vol. 24, No. 9, pp. 487- 492, September 2018.
(ISSN 2383- 6318)
*Jaehyun An and Youngjoong Ko "Semi- automatic Expansion for a
Chatting Corpus Based on a K- means Clustering Method And Similarity
Measure" Journal of Korean Institute of Information Scientists and
Engineers (KIISE), Vol. 46, No. 5, pp. 440- 447, May 2019. (ISSN
2383- 630X)
*Hongyeon Yu, Jaehyun An, Jeongmin Yoon, Hyemin Kim and Youngjoong Ko
- 2 -
"Simple Methods to Overcome the Limitations of General Word Representations
in Natural Language Processing Tasks" Computer Speech and Language,
Elsevier Science, Vol. 59, pp. 91- 113, January 2020.(SCIE, ISSN 0885- 2308)
<논문>
*Jaehyun An, Hokyung Lee and Youngjoong Ko. Korean Dependency
Relation Labeling Using Bidirectional LSTM CRFs Based on the
Dependency Path and the Dependency Relation Label Distribution of
Syllables. Proceedings of the 28th Annual Conference on Human and
Cognitive Language Technology (HCLT 2016), pp. 14- 19, October 2016.
*Hyemin Kim, Jungmin Yoon, Jaehyun An, Kyoungman Bae and
Youngjoong Ko. Syllable- based Korean POS Tagging using POS
Distribution and Bidirectional LSTM CRFs. Proceedings of the 28th
Annual Conference on Human and Cognitive Language Technology
(HCLT 2016), pp. 3- 8, October 2016.
*Jaehyun An and Youngjoong Ko. The Korean Dependency Parser for
Pointer Networks based on Multi- task learning and Relation Label
Distribution of Syllables. Proceedings of Korea Software Congress 2017
(KSC 2017), pp. 613- 615, December 2017. (우수논문발표상 수상)
*Jaehyun An and Youngjoong Ko. Semi- automatic Expansion for a
Chatting Corpus Based on Similarity Measure Using Utterance
Embedding by CNN. Proceedings of the 30th Annual Conference on
Human and Cognitive Language Technology (HCLT 2018), pp. 95- 100,
October 2018. (우수논문상 수상)
<프로젝트 및 연구>
* May 2016 January 2017, Principal Investigator (PI) involved in
“WiseWordNet 확장을 위한 사전 Entity Linking 연구 (Entity Linking for
Expanding WiseWordNet).” Supported by ExoBrain Project, 한국전자통신
연구소 (ETRI). (IS lab. Dong- A University)
*April 2017 December 2019, Co- Principal Investigator (Co- PI) involved in
“기계학습용 텍스트 데이터 레이블 자동생성 및 검증도구 개발
(Development of Automatic Text Data Labeling and Verification Tools for
Machine Learning).” Supported by 정보통신/방송 기술개발사업, 미래창조
과학부 (Ministry of Science, ICT and Future Planning) . (IS lab. Dong- A
University)
<특허>
*Youngjoong Ko et al. Method for Natural Language Understanding
Based on Distribution of Task- specific Labels.출원일: 2017.08.18 (출원번호 제 10- 2017- 0104871 호)
*Youngjoong Ko et al. System and Method for Expansion Chatting Corpus Based on Similarity Measure Using Utterance Embedding by CNN. 출원일: 2018.10.05 (출원번호 제 10- 2018- 0119113 호)
<저작권 / 판권>
*Youngjoong Ko et al. 위키피디아를 이용한 Entity Linking(개체연결) 시스템.
창작일: 2017.01.23, 공표일: 2017.01.25 (C- 2017- 003030)
*Youngjoong Ko et al. 레이블 자동 생성을 위한 한국어 의존 구문 분석기.
창작일: 2017.11.7, 공표일: 2017.11.7 (C- 2017- 029032)
<수상경력>
* 2016 국어경진대회 일반분야 동상.
* 2017 KSC 우수논문발표상.
* 2018 HCLT 우수논문상.
세부분야*Machine Learning, Deep Learning
*Natural Language Processing(NLP)