Sumin Han
last updated: Feb 21st, 2024
Sumin Han, Ph.D. Candidate – Expected Graduation Date: August, 2024, CDSN Lab (Advisor: Dongman Lee)
School of Computing, Korea Advanced Institute of Science and Technology (KAIST)
Education
- 18’~ Integrated master’s/doctoral program (doctoral), KAIST - School of Computing
- 18’ B.S. in KAIST - School of Computing, Advanced Major
- 13’ Korea Science Academy of KAIST
Research Interests
Urban Computing, Spatio-temporal Deep Neural Network, Road2Vec/Region2Vec, Urban Simulation
Publications
International Publications
2024
- Sumin Han, Youngjun Park, Sonia Sabir, Jisun An, and Dongman Lee, “Improving Real Estate Appraisal with POI Integration and Areal Embedding,”, Arxiv Code, DMO-FinTech Workshop held in conjunction with the PAKDD conference in May 7-10, 2024.
- Youngjun Park and Sumin Han. “Enhancing Population Predictions in Developing Area through Building
Construction Data”. In: Association of European Schools Of Planning (AESOP) (2024)
2023
- Sumin Han, Youngjun Park, Minji Lee, Jisun An, and Dongman Lee, “Enhancing Spatio-temporal Traffic Prediction through Urban Human Activity Analysis,” 32nd ACM International Conference on Information and Knowledge Management (acceptance rate 24%) Paper Code
- Youngjun Park, Sumin Han, Jisun An, and Dongman Lee “Encoding Urban Trajectory as a Language: Deep Learning Insights for Human Mobility Pattern”, The paper will be presented at the Association of Collegiate Schools of Planning (ACSP) conference 2023 in Chicago, IL. Paper Code
2020
- Sumin Han, Dasom Hong, and Dongman Lee, “Exploring Commercial Gentrification using Instagram Data”, The 2020 IEEE/ACM International Conference on Advances in Social Network Analysis and Mining (ASONAM), December. 2020. link
- Sumin Han, Kinam Park, and Dongman Lee, “Discovering Daily POI Exploitation using LTE Cell Tower Access Traces in Urban Environment”, International Conference on Social Informatics 2020, October. 2020. link
- Dongmin Kim, Sumin Han, Heesuk Son, and Dongman Lee, “Human Activity Recognition using Semi-Supervised Multi-Modal DEC for Instagram Data”, PAKDD 2020. link
Domestic Publications
2024
- Doyoon Park, Sumin Han, Youngjun Park, Dongman Lee, “Geographical Visualization Interface of Human Travel Path to Understand Human Mobility”, Paper DEMO
2022
- 김창희, 한수민, 이동만, 지하철 이용패턴을 통한 역단위 장소성 추론, KCC 2022
- Cao Long, Sumin Han, Dongman Lee, 인스타그램 게시물의 장면 그래프 적용을 통한 멀티모달 지도학습 기반 인간 활동 인식율 향상, KCC 2022
2021
- 김수빈, 한수민, and 이동만. “서울시의 토지 이용에 따른 도로의 차량 속도의 상관관계 분석.” 한국정보과학회 학술발표논문집
(2021): 119-121. link
- Marciniak, Zofia, Sumin Han, and Dongman Lee. “Understanding the impact of the Weather on Human Mobility via LTE Access Traces in Seoul Districts.” 한국 HCI 학회 학술대회
(2021): 615-618. link
2020
- 정종구, 한수민, and 이동만. “인스타그램에서 나타나는 인간 활동 패턴의 연간 변화 분석(Analysis of Annual Changes in Human Activity Patterns on Instagram).” 한국 HCI 학회 학술대회
(2020): 723-726. link
- 안정미, 한수민, and 이동만. “인스타그램 데이터를 활용한 장소성 분석 및 시각화(Visualizing a Sense of Place Using Instagram Data).” 한국 HCI 학회 학술대회
(2020): 929-933. link
2018
- 한수민, and 김명호. “베이지안 개인화 순위 방법 기반 대용량 이종 네트워크에서의 효율적인 랜덤 워크 순위 계산(Random Walk with Restart on a Large-scale Heterogeneous Network Based on the Bayesian Personalized Ranking).” 한국정보과학회 학술발표논문집 (2018): 889-891.APA link
Patents
- 이동만, 한수민, “다중지역특징 인식 시공간 대중교통 예측 시스템 및 그 방법 (SYSTEM AND METHOD FOR PREDICTING MULTI-LOCAL FEATURE RECOGNITION SPATIO-TEMPORAL PUBLIC TRANSPORT)”, 출원번호: 10-2023-0055885 (2023-04-28)
- 이동만 한수민, “도시 환경에서 LTE 셀 타워 액세스 트레이스를 이용한 일일 POI 이용 발견 방법 (Method of Discovering Daily POI Exploitation Using LTE Cell Tower Access Traces in Urban Environment)”, 출원번호: 10-2021-0047046 (2021-04-12), 등록번호: 10-25947500000 (2023-10-24)
- 이동만 김동민 한수민 손희석, “소셜 미디어 데이터에 대한 준 지도 멀티 모달 딥 임베디드 클러스터링을 이용한 인간 활동 인식 방법 (Method for Human Activity Recognition Using Semi-supervised Multi-modal Deep Embedded Clustering for Social Media Data)”, 출원번호: 10-2021-0009319 (2021-01-22), 등록번호: 10-25149200000 (2023-03-23)
Experiences
- Deep Urban: https://deepurban.kaist.ac.kr/ (Jan, 2023 ~)
- Satrec Initiative Co., Ltd. (Jul 2017 - Aug 2017 · 2 mos)
- Parallel Computing Framework Design
- Skills: Parallel Computing · MPI · CUDA · docker · CentOS
- 로톡 LAWTALK (Mar 2014 - May 2014 · 3 mos): Software Engineer
- Full-time SW developer
- Skills: JavaScript · HTML · PHP · Git · Node.js · MongoDB · Vim · Express.js · AngularJS · Agile Methodologies · Firebase
Curriculum Vitae
Hobbies
Cooking, Chess (sucern@chess.com)
Favorite Books