My life goal is to lower the barriers of knowledge through data. I am currently working for Naver Papago – and lowering the barriers of linguistic knowledge by building machine translation models. Current areas of interest include context-aware machine translation, and translation evaluation.
You can often find me on the Web as
- 2016-Present: Machine learning scientist, Naver Corporation
- 2009-2016: M.S., Ph.D., Data Mining Lab, Seoul National University
- 2004-2009: B.S., Industrial Engineering / Economics (minor), Seoul National University
- 2020-06: We had a chat with Kyunghyun Cho at TensorFlow Korea. It’s not everyday you meet your role model…
- 2020-05: Read this blog to see how Papago MT and NLE MT group are contributing to research for COVID-19. (Related news articles)
- 2019-10: We are having the second NLP-OSS workshop, co-located with EMNLP 2020. Consider submitting a paper!
- 2019-06: Here’s an article about me developing Papago’s new honorific translation feature. (In Korean)
- 2019-04: Interested in machine translation for Asian languages? Come see our interview with Slator.
- 2019-01: We had an interview with Data Skeptic regarding KoNLPy! Go have a look.
- 2018-01: I’m organizing Deep Learning Camp Jeju 2018, a sequel of the ML Camp Jeju 2017.
- 2017-11: I’m organizing the inaugural NLP-OSS Workshop with Masato Hagiwara, Dmitrijs Milajevs, Liling Tan – a workshop devoted to open source software regarding NLP, to be co-located with ACL 2018.
- 2017-06: I’m organizing and participating in the Machine Learning Camp 2017 at Jeju. I will be mentoring Judit Ács, and we will be building a morphology segementation model for Hungarian!
- 2016-12: My team at Naver, Papago, won 1st place in Naver’s 2016-4Q Project Presentations, and 2nd place among all projects presented at 2016 Naver User Connect Day!
- 네이버 서비스위원회 과제발표회 2016년 4분기 성과부문 1위, 네이버 사용자 커넥트 데이 2017년 성과부문 2위.
- 2016-08: I’ve (finally) graduated Seoul National University as a Ph.D., and joined the machine translation team at Naver Corporation to build Papago!
- 2020: Program commitee for ACL, EMNLP, AACL-IJCNLP 2020. Co-organizer of 2nd Workshop for Natural Language Processing Open Source Software (NLP-OSS) @EMNLP 2020.
- 2018: Co-organizer of Workshop for Natural Language Processing Open Source Software (NLP-OSS) @ACL 2018.
- 2017-Present: Committee member at TensorFlow Korea
- 2011-2014: Data miner at Team POPONG
- Spring 2015: Data Mining @SeoulTech, Lecturer Website
- Spring 2015: Business Analytics @SeoulTech, Lecturer Website
- Fall 2012: Data Mining @SNU, Teaching assistant
- Fall 2012: Engineering Research Practice 1 @SNU, Teaching assistant
- I’m sharing experiences for the Web, books, music, quotes, and movies.
- Personal characteristics:
- Friends call me a dreamer or romantist to a fault - but I enjoy living my life as such.
- I am a productivity geek.
- I spend much of my free time automating things.
- I easily get addiicted to things, so I don’t start things easily.
- Things I love:
- Riding my bike, swimming, scuba diving, playing the piano and making my own fresh coffee with a siphon.
- Adventures, smiling, learning, and pursuing community action.
- Exploration! As opposed to exploitation. I particularly enjoy exploring the realm of cuisines. I have two particular dishes I do not enjoy – 홍어, 추어탕 – but who knows? I may even come to like these in the future.
- Openness, sharing, and cooperation (ex: Everyware)
- I live with a cat named Lea.
- These are some things I believe in:
- Life is about the journey, not about the destination.
- Languages themselves are not goals but tools of communication.
- Happiness brings smiles, and smiles also bring happiness.
- That there is no wrong and no right, is right.
- The individual is not something complete, perfect or finished, but something moving, changing, and above all initiating instead of final.
- You can make money without doing evil.
- Todo lo que es tiene más razones para ser que para no ser.
- Treat those around you as reasonable people, and when they have different opinions, try to discover what problem they are solving.
- Explicit is better than implicit.