Data-Driven Decision Science Lab
Better Decisions by Better Data and Methodology
The D3SL conducts cutting-edge research at the intersection of statistics, data science, and privacy-preserving technologies. Our work aims to advance statistical methodology for the safe and efficient use of data in both public and private sectors, particularly focusing on complex survey data, incomplete data, and privacy-constrained environments.
People
Faculty
PhD Course
Jeong, D.
Synthetic Data & Data Privacy
Kim, C.
Imbalanced data & Imputation
Kim, I.
Survey Sampling / Political Engineering
Oh, H.
Missing Data Analysis
Kim, J.
Anomaly Detection
Master Course
Lee, C.
Time-series data synthesis
Park, K.
Text data analysis
Park, J.
Data synthesis
Chun, S.
Data synthesis
Jeon, J.
Causal analysis
Research
Manuscripts (Submitted)
- A Survey on Tabular Data Synthesis: Generation, Evaluation, and Benchmarking Experiments
- Imputation-Based Causal Analysis for Observational Data
- Beyond the Privacy-Utility Trade-off: Cryptographic Federated Calibration for Enhanced Model Performance
- Auditing Alignment: A Framework for Measuring Value Drift in Public Sector AI Procurement
- Bias-corrected estimation in causal mediation analysis
Published Works (Statistics & Data Science, Other Domains)
- (2025) GAM-MIDAS: Generalized additive model-based mixed-data sampling regression with informal data, Applied Soft Computing
- (2025) Calibrated Mixup for Imbalanced Regression on Tabular Data, Pattern Recognition
- (2025) Re-sampling Calibrated SNN Loss: A Robust Approach to Non-IID Data in Federated Learning, Expert Systems
- (2025) Identification enhanced generalised linear model estimation with nonignorable missing outcomes, Journal of the Royal Statistical Society: Series A
- (2025) Optimizing Federated Learning: Addressing Key Challenges in Real-World Applications, IEEE Internet of Things Journal
- (2025) A Generalized Theory of Mixup for Structure-Preserving Synthetic Data, AISTATS
- (2025) 선거 여론조사의 품질과 대표성에 관한 분석, 제20대 제21대 국회의원선거를 중심으로, 조사연구
- (2024) Overcoming Data Imbalance in Federated Learning with Calibration Weighting, IEEE International Conference on Big Data
- (2024) Online News-Based Economic Sentiment Index, IEEE Transactions on Big Data
- (2024) Robust demand estimation with customer choice-based models for sales transaction data, Production and Operations Management
- (2024) Proposing Bayesian Hierarchical Growth Curve Models (BHGCMs) for Tourism and Hospitality Research, International Journal of Hospitality & Management
- (2024) Relabeling & Raking Algorithm for Imbalanced Classification, Expert Systems with Applications
- (2023) Resampling Approach for One-Class Classification, Pattern Recognition
- (2023) Quantile regression with multiple proxy variables, STAT
- (2023) A new global measure to simultaneously evaluate data utility and privacy risk, IEEE Transactions Information Forensics and Security
- (2023) Potential threat of microplastics to humans: toxicity prediction modeling by small data analysis, Environmental Science: Nano
- (2022) Confirmatory aspect-level opinion mining processes for tourim and hospitality research: a proposal of DissBUS, Current Issues in Tourism
- (2022) Data integration of National Dose Registry and survey data using multivariate imputation by chained equations, Plos One
- (2022) Cosine-based variable bandwidth selection for nonparametric spectral density estimation under long-range dependence, Journal of Statistical Computation and Simulation
- (2022) A small-data-driven model for predicting adsorption properties in polymeric thin films, Chemical Communications
- (2021) A note on stationary bootstrap variance estimator under long-range dependence, Statistics & Probability Letters
- (2021) Integration of statistical and administrative agricultural data from Namibia, Statistical Journal of the IAOS
- (2021) A growth curve-based Bayesian hierarchical model for multi-building energy use data analysis, Building and Environment
- (2021) COVID-19, social distancing, and risk-averse actions of hospitality and tourism consumers: A case of South Korea, Journal of Destination Marketing & Management
- (2021) COVID-19: Were public health interventions and the disclosure of patients' contact history effective in upholding social distancing? Evidence from South Korea, Journal of Multidisciplinary Healthcare
- (2020) Does the written word matter? the role of uncovering and utilizing information from written comments in housing ads, Journal of Housing Research
- (2020) An assessment of opinions and perceptions of smart thermostats using aspect-based sentiment analysis of online reviews, Building and Environment
- (2020) A least squares-type density estimator using a polynomial function, Computational Statistics & Data Analysis
- (2019) Frequency domain bootstrap for ratio statistics under long-range dependence, Journal of the Korean Statistical Society
- (2019) Impacts of fractional hot-deck imputation on learning and prediction of engineering data, IEEE Tran. Knowledge and Data Engineering
- (2019) Cost-effective extreme case-control design using a resampling method, Evolutionary Bioinformatics
- (2018) Proposing a missing data method for hospitality research on online customer reviews: An application of imputation approach, International Journal of the Contemporary Hospitality Management
- (2018) FHDI: An R package for fractional hot deck imputation, R Journal
- (2017) Multiple imputation for nonignorable missing data, Journal of the Korean Statistical Society
- (2017) Correlation estimation with singly truncated bivariate data, Statistics in Medicine
- (2017) Energy efficiency in US residentail rental housing: Adoption rates and impact on rent, Applied Energy
- (2016) A post-hoc genome-wide association sutdy using matched samples, International Journal of Data Mining and Bioinformatics
- (2016) A propensity-score-adjustment method for nonignorable nonresponse, Journal of Survey Statistics and Methodology
- (2015) Two-phase sampling approach to fractional hot deck imputation, Proceedings of the Survey Research Methods Section
- (2014) Propensity score adjustment with several followups, Biometrika
Project
국가 R&D
연구용역 (최근 4년)
Alumni (최근 5년)
학계
- Yonsei University (Post Doctor)
- Carnegie Mellon University (PhD Student)
- University of Texas, Austin (PhD Student)
산업
- 나이스평가정보
- 대한항공
- 리디
- DB 화재
- LG 전자
정부&공공
- 금융감독원
- 주택도시보증공사
- 정보통신정책연구원
- 한국은행