Hyunyoung Lee |


Summary


I've always tried to take an effort to improve my software skill. Apart from education at university, I got experience a variety of software development. developing at Natural Language Processing Lab, SAMSUNG SOFTWARE MEMBERSHIP and then I developed a new technology of SSD like NVMe and LightNVM and so on. Currently, I'm looking for job that I can work for thechnology that I can have interest in.

Open-source contribution


Merged request and issue

Open

Closed and not merged

Development Experience


Natural Language Processing lab in Kookmin University | Seoul, South Korean | 2017.06 --- present

Researcher

  • have worked as tech leader and researcher and engineer in Natural Language processing labs
  • researched and developed “Korean Dependency Parsing by using Transiion-based Deep Learning Techonology” with Tensorflow. [Slide-None]. [URL-None].
  • researched and developed “Compound Noun Decomposition by using Bi-LSTM and Linear-chain CRF” with Tensorflow. [Slide-None]. [URL-None].
  • researched and developed “Bi-LSTM-CRF and Syllable Embedding for Automatic Spacing of Korean Sentences” with Tensoflow. [Slide-None]. [URL-None].
  • developed and carried out python open source project for Korean Natural Language toolkit. I have a role as Tech Leader and designed the total structure about the Open source which is how to contribute and test with test module, automatic build system, code checking, and documentation. I also made guide website with readthedocs and sphinx for python platform. [KoNLP websit], [KoNLP Github].
  • researched and developed "Deep learning Method of Keyword Generation by using Doc2Vec and Word2Vec" - Award : 2nd Prize in Poster Presentation(IC-LYCS2018), Feb 2018 by Asia Pacific Society for Computing and Information Technology. [Poster-None].
  • researched and developed "filtering Spam Text by using Sen2Vec and FeedFoward Neural Network" with Tensorflow, SVM Light, Google Word2Vec(C source). [Slide]. [URL].
  • developed the recommendation word in text which is A B X C D, A B X, or X A, B. In the case, What is the best word on the position, X. - Toy Project [Github].
  • Seminar of NLP technologies with deep learning
    1. “The Curious Case of Neural Text Degeneration (2021.07.09)”. [Slide].
    2. “Presentation About NLP pretraining for embedding on Applied web information processing class (2021.05.17)”. [Slide].
    3. “Cross-lingual Language Model Pretraining (2021.05.07)”. [Slide].
    4. “MASS: Sequence to Sequence Pre-training for Language Generation (2021.04.03)”. [Slide].
    5. “BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension (2021.03.04)”. [Slide-Long Ver.]. [Slide-Short Ver.].
    6. “Retrofitting word vector to semantic lexicons (2020.01.08)”. [Slide].
    7. “Learned in Translation: Contextualized Word Vectors(2019.11.20)”. [Slide].
    8. “Language Grounding and Never-ending Language learning. (2019.06.25)”. [Slide].
    9. “GloVe: Global Vectors for Word Representation(Paper seminar). (2019.05.31)”. [Slide].
    10. “Oversampling Scheme using Conditional GAN(Paper seminar).(2018. 11. 07)”. [Slide].
    11. “Implementation of Korean Sentence Similarity using Sent2Vec sentence embedding(Paper Seminar).(2018. 10. 19)”. [Slide].
    12. “Distinguishing Japanese Non-standard Usage from Standard ones(Paper Seminar).(2018.10.05)”. [Slide].
    13. “Universal Dependencies For Finnish(Paper seminar).(2018.09.28)”. [Slide].
    14. “Visual Question Answering(VQA).(2018.07.09, 2018.07.16)”. [Slide].
    15. “Tensorflow Practice For NLP open seminar.(2018.05.25)”. [Slide].
    16. “Skip-gram & CBOW(continuous bag-of-words).(2017.11.28)”. [Slide].
    17. “Word2Vec & Language Model.(2017.10.23)”. [Slide].
  • Research Projects
    1. AI 드라마 창작의 최신 기술동향 및 전망에 대한 연구 by SBS 미디어 홀딩스 주식회사 2020. 03. 15 ~ 2020. 08. 31
    2. Modeling News Credibility Prediction with Computational Methods(Kor ver: 공하적 분석기법을 활용한 기사 신뢰도 예측 모형 연구) by NRF(Kor ver. 한국연구재단) 2018. 07. 01 ~ 2019. 06. 30
    3. Automatic Spelling Correction and Abstractive Text Summarization including Rare Words by using Sequence-to-sequence RNNs(Kor ver. 순차정보 학습 순환 신경망을 이용한 오류어 자동 교정 및 회소 단어가 포함된 텍스트의 자동 요약) by NRF(Kor ver. 한국연구재단) 2017. 09. 01 ~ 2020. 05. 31

SK Hynix memory solutions | San Jose, CA | 2016.02 --- 2017.01

Software Engineer(Intership)

  • worked in Advanced storage team for NVMe protocol and SSD’s new technology.
  • carried out “Robusta” project which transfers features of SSD from firmware to Linux Host, based on NVMe protocol. The features of hardware is Translation Map, wear-leveling, Garbage Collection.
  • tested and researched rrpc(Round-Robin page-based hybrid FTL) and pblk(physical block device) of LightNVM on QEMU-NVMe(hypervisor) with FIO.
  • tested and researched Application-Managed FTL and Device-mapper with FIO, I also investigated about Multi-Queue layer on Linux.
  • tested SSD(DumbSSD) for Robusta with NVMe-cli Open Source tool(ioctl), changing the OpenSource tool.
  • investigated and researched F2FS which is friendly filesystem of SSD and RocksDB for Robusta.
  • tested and verified all flash array system for performance of NVMe device on LINUX with CPU-affinity, CPU-isolation and IRQ handler. the all flash array system is storage that all consists of SSD.
  • have verified performance of filesystem on SSD with FIO, filebench tool and postmark tool, tuning NVMe-cli to get a WAF, file systems that I verified are F2FS, ALFS, ext4. [GitHub].

SAMSUNG SOFTWARE MEMBERSHIP | Seoul, South Korea | 2014.01 --- 2015.06

Software Membership

  • carried out a variety of projects in the group that Samsung Electronics supports, this group consists of undergraduates from a variety of Universities.
  • developed windows(UI), Server(database) and a dish-recognition algorithm of the application, “Smart Refrigerator tell the location of the dish to a person who is visually impaired”, with Samsung Electronics' Home Appliances Business Department for designing new IOT product, using C#(Windows), Emgu CV, MySQL, Open Source(Clipper).
  • developed windows(UI), USB communication module and graphic card’s algorithm to adjust monitor's brightness, contrast and gamma of Windows application, “Monitor Calibration using smart phone device”, using C#(windows), Java(Android).
  • developed algorithm of battery-efficiency and tested the performance of project, “Customizing Governor in big.LITTLE Architecture”, in order to improve efficiency of smart phone’s battery using C/C++(Linux), Java(Android).
  • developed a sever of application, ”User Customizing the English wrong note ”, using Apache, PHP, MySQL, JavaScript, HTML, CSS, Java(Android).
  • developed algorithm of Lego Mindstorms for Leg robot’s war.
  • taught about “ the basic algorithm & data structure”, to co-work.
  • worked as advisor about project especially involved natural language processing(nlp) and Linux when co-work started such a project.
  • learned about communication skill, ability of planning, Leadership, while I organized team and planned to do project as leader or member of team.

Natural Language Processing Labs in Kookmin University | Seoul, South Korea | 2013.04 --- 2014.12

Internship(undergraduate research assistant)

  • wrote the paper, “Plagiarism Detected Source Retrieval and Text Alignment” that posted in National Assembly Library Journal and accepted in 26th Annual Conference on Human and Cognitive Language Technology(HCLT-2014).
  • developed UI and algorithm of project, “Plagiarism Detected Source Retrieval and Text Alignment” using Python.
  • developed and designed the Knowledge-Base of Automatic Scoring of Korean Short(Clause & Phrase) Answer using MySQL, C/C++.
  • developed a PDF document filter using C/C++. - Toy Project
  • carried out project, “CopyCheck : A Two-step Approach to Plagiarism Detection”, using C/C++.
  • learned natural language processing and information Search skills, trained in coding and writing.

Korea IT Volunteers | Phrae, Thailand | 2012.07 --- 2012.08

Volunteer & Teacher

  • volunteered in RongFong ICT Center at the Phrae, Thailand for teaching IT & Korea's culture to Thais, RongFong Villagers & NAMCHANG Elementary School's students.
  • learned communication skill with foreigners who can't speak English at all while working and teaching.
  • learned problem-solving ability while resolving unexpected problems in this volunteers.
  • Korea IT Volunteers is hosted in NIA(NATIONAL INFORMATION SOCIETY AGENCY) & ITU(International Telecommunication Union).

Kookmin university | Seoul, South Korea | 2008.03 --- 2016.08

Student

  • developed term project, “Visualizing Bigdata”, using sigma library in Javascript.
  • carried out term project of encryption & decryption using AES algorithm and analysis of packet using the Wireshark.
  • developed term project, “Mini-C compiler using Lex”, Yacc, C.
  • carried out term project, “Robot cleaner algorithm with Backtracking” using C.
  • developed term project, “Cover flow” using openGL.
  • carried out project, “line Tracer with Lego Mindstoms” using C.

Education


Kookmin University | Seoul, South Korea | 2019.09.02 --- Present

Ph.D. candidate in Computer Science, Field of Study: Computer Science, in particular, Natural Language Processing
GPA - 4.33 / 4.5, 98.1 % cumulative average

  • ongoing
  • worked as TA(teaching assistant) at Python(2019 2nd, 2020 1st, 2021 1st) class for students who don't major in CS and engineering.
  • Toy projects for autonomous vehicle and Visual SLAM and Traffic sign classification
    1. German traffic sign classicfication in Autonomous Driving Recognition and Decision class[Github]
    2. Project 1 for Perception and Planning in Advanced Vehicle Intelligence class [Github]
    3. Project 2 for Viual SLAM in Advanced Vehicle Intelligence class [Github]
  • On paper seminar on my classes
    1. (team presentation)SSD: Single Shot MultiBox Detector in European conference on computer vision 2016 in Autonomous Driving Recognition and Decision class (2021. 05.25)[Slide].
    2. ADA: Adaptive Deep Log Anomaly Dectector in Proceedings of IEEE INFOCOM 2020 on Advanced Topics in Computer Science class (2020.11.04)[Slide].
    3. Real-time Personalization using Embeddings for Search Ranking at Airbnb in Proceedings of KDD 2018 on Data Mining class (2019.12.13)[Slide].
    4. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding In Proceedings of NAACL 2019 on Selected Topics in Computer Science class (2019.12.02)[Slide].
    5. Neural Machine Translation Inspired Binary Code Similarity Comparison beyond Function Pairs In Proceedings of NDSS2019 on Advanced Information Security class (2019.10.24)[Slide].
  • On Presentation on my classes
    1. Drama Script's Genre Classification and Keywords Clustering on Advanced Information Computing class(2020.07.03)
    2. Sentiment analysis based on LSTM (long short-term memory) on Introduction to Deep learning class (2020.06.29)
    3. Go Green Insurance Migration on Cloud Architect by AWS Academy class (2020.06.30)[Slide].

Kookmin University | Seoul, South Korea | 2017.08.28 --- 2019.08.21

Master of Engineering(M.Eng.) in Computer Science, Field of Study: Computer Science, in particular, Natural Language Processing
GPA - 4.43 / 4.5, 99.2 % cumulative average

  • On paper seminar on my classes
    1. Memory Networks In Proceedings of ICLR 2015 on Selected Topics in Computer Science class (2018.11.21)[Slide].
    2. Large-Scale and Language-Oblivious Code Authorship Identification In Proceedings of CCS'18 on Advanced Topics in Computer Science class (2018.11.06)[Slide].
    3. DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning In Proceedings of CCs'17 on Advanced Information Security class (2017.12.06)[Slide].
  • On Presentation on my classes
    1. Word Embedding Method of SMS Messages for Spam Message Filtering porject on Machine learning class (2019.04.16)[Slide].
    2. Korean Language Autospacing team project in class of Advanced Tech. in Machine Learning (2018.06.12)[Slide][Github].
    3. How to use Tensorflow on Advanced Information Computing class (2018.04.30)[Slide].
    4. Word embedding on Advanced Information Computing class (2018.03.19)[Slide].
    5. Interprocess communication (What Middleware layer is about socket communication with TCP and UDP) on Advanced Distributed System Class (2017.10.31)[Slide].
  • worked as TA(teaching assistant) at Python(2017, 2nd) class and Excel class & Codly(2018, 1st) class like block coding(scratch), JavaScript(2018, 2nd) class, Excel(2019, 1st) class for students who don't major in CS and engineering.
  • worked as TA(teaching assistant) at Information retrieval & data mining(2017, 2nd) class.
    1. What is the NLTK(Natural Language Processing Toolkit)(2017.11.28)[Slide].
    2. SVM-light and SVM-multiclass Practics(2017.11.16)[Slide].

Kookmin University | Seoul, South Korea | 2008.03.03 --- 2016.08.24

Bachelor of Engineering in Computer Science
GPA - 3.69 / 4.5, CS GPA - 3.67 / 4.5, 90.7 % cumulative average

  • developed my Git-Hub pages, static website, using Jekyll, html, css, git.
  • won scholarship eight times.
  • won the 3rd Award in the ACM-ICPC of Kookmin University
  • studied a range of the basic Computer Science and finished more than 7 term projects.

MILITARY SERVICE


Republic of Korea Army(Military Service) | Dongducheon, Gyeonggi Province, South Korea | 2009.03 --- 2011.01

Soldier at HR & Financial

  • worked on things related to Accounting and HR Service as a personnel & accounting clerks in Army's Human Resources.
  • wrote and arranged lots of documents using MS office and Hancom tool.
  • prepared mobilization training as director leading lots of soldier.
  • worked on thing related to military law.
  • learned documentation skill using a variety of tools, leadership and accounting & HR tasks.

ACTIVITIES & Societies


PAPERS


JOURNAL

JOURNAL in Korea

CONFERENCE PRESENTATIONS

CONFERENCE PRESENTATIONS in Korea

AWARDS & HONORS


  • Best Paper Bronze at International Conference on Smart Media and Applications (SMA 2020) | Jeju, South Korea | 2020.09
  • Excellence Award at Automatic Spacing and Compound Noun Decomposition: 2018 Korean Natural Language Processing Contest | Seoul, South Korea | 2018.10.12
  • 2nd Prize in Poster presentationon, at 2018 International Conference for Leading and Young Computer Scientists (IC-LYCS2018) (by Asia Pacific Society for Computing and Information Technology) | Okinawa, Japan | 2018.02.11
  • Encouragement Prize(The Electronic Times CEO Prize), The 27Th Global Software Contest | Seoul, South Korea | 2015.12.16
  • The Third Award, 2015 ACM-ICPC Asia-Daejeon Korea National First Round in the Kookmin University | Seoul, South Korea | 2015.10.05
  • The First Award(1st Award), IOT Competition supported by Samsung Electronics' Home Appliances Business Department | Suwon, Gyeonggi-do, South Korea | 2015.05.21
  • Encouragement Prize, 2014 Engineering Education Festival | Seoul, South Korea | 2014.11.07
  • Encouragement Prize, 2013 Engineering Education Festival | Seoul, South Korea | 2013.11.15
  • Encouragement Prize, 2013 Korean Language Information Processing System Competition | Seoul, South Korea | 2013.10.11

TECH. SKILL


  • Programming: (Intermediate) C, C++ , Python, (Low) Shellscript, C#, PHP, Java, JavaScript, perl, LISP.
  • Development Tools: MS Visual Studio and Eclipse, Sublime Text, Vi editor.
  • Embedded : (Intermediate)Raspberry Pi, ODROID-XU+E & A.
  • Computer Graphics : (Intermediate)Window DC, (Low)OpenGL.
  • Platform : (Intermediate) Windows, Linux(Ubuntu,Centos7), (Low) Android.
  • Algorithm: (Intermediate) Basic Data Structure, Graph, Greedy, Searching(DFS, BFS), Sorting, Dynamic Programming, Tree.
  • Natural Language processing(Intermediate).
  • SSD : NVMe, QEMU-NVMe, OpenChannelSSD(LightNVM), FIO, NVMe-Cli, filebench tool.
  • Machine Learning platform: Tensorflow.