ABOUT ME

I am an Assistant Professor of Statistics at Sookmyung Women's University and an Adjunct Professor of Quantitative Applied Economics at Sungkyunkwan University.

Before joining SMU, I served as a Senior Economist in the Office of Economic Modeling and Policy Analysis at the Bank of Korea for 14 years.

I completed my Ph.D. in Statistics at Pennsylvania State University, focusing on interpretable statistical learning, advised by Dr. Jia Li, and hold a B.A. in Economics and Statistics from Korea University.

Research Interests

My research interests span a range of topics, including:

• Interpretable neural networks and machine learning,
• Big data analytics for social science, economics, and business,
• Text-related statistical modeling, High-dimensional data mining, Nonparametric methods.

PUBLICATIONS

* Corresponding author

Peer-reviewed Journal Articles

( International Journals : SCIE/SSCI )

• Lee, Y.J., Kim, S., Seo, B., & Park, K.Y. (2024+). Measuring Monetary Policy Surprises Using Text Mining: The Case of Korea. Will be submitted soon (the manuscript is available upon request).


• Seo, B.* (2024+). Econometric Forecasting Using Ubiquitous News Text: Text-enhanced Factor Model. International Journal of Forecasting. In press.
  ^{[SSCI, Top 3.1% Economics][code]}


• Seo, B.* & Li, J. (2024). Explainable machine learning by SEE-Net: closing the gap between interpretable models and DNNs. Scientific Reports, 14, 26302.
  ^{[SCIE, Q1 Multidisciplinary][pdf][code]}


• Seo, B.*, Lee, Y., & Cho, H. (2024). Measuring News Sentiment of Korea Using Transformer. Korean Economic Review, 40(1), 149-176.
  ^{[SSCI, Economics][pdf]}


• Seo, B.*, Lin, L., & Li, J. (2022). Mixture of Linear Models Co-supervised by Deep Neural Networks. Journal of Computational and Graphical Statistics, 31(4), 1303-1317.
  ^{[SCIE, Q1 Statistics][arXiv][code]}


• Seo, B.*, Lin, L., & Li, J. (2021). Block-wise Variable Selection for Clustering via Latent States of Mixture Models. Journal of Computational and Graphical Statistics, 31(1), 138-150.
  ^{[SCIE, Q1 Statistics][code]}


• Li, J., Seo, B., & Lin, L. (2019). Optimal transport, mean partition, and uncertainty assessment in cluster analysis. Statistical Analysis and Data Mining: The ASA Data Science Journal, 12(5), 359-377.
  ^{[SCIE, Q1 Statistics][code]}

( Domestic Journals : KCI )

• Shin, S.J., & Seo, B.* (2024). Real-Time Private Consumption Prediction Using Big Data. (in Korean). The Korean Journal of Applied Statistics, 37(1), 13-38.


• Yoo, J.K., Park, Y., & Seo, B.* (2024). Using Noise Filtering and Sufficient Dimension Reduction Method on Unstructured Economic Data. (in Korean). The Korean Journal of Applied Statistics, 37(2), 119-138.

Working Papers

• 서범석, (2024). 빅데이터 기반 민간소비 패턴 분석과 전망[Private Consumption Pattern Analysis and Forecasting Based on Big Data]. BOK이슈노트, No.2024-24.


• Seo, B.*, Lee, D., & Cho, H. (2023). Point and Risk estImation using an enSemble of Models for Nowcasting: PRISM-Now. BOK Working Paper, No.2023-27.
  
  


• Seo, B., (2023). Econometric Forecasting Using Ubiquitous News Text: Text-enhanced Factor Model. BOK Working Paper, No.2023-10.


• 서범석, (2023). AI 알고리즘을 이용한 산업모니터링: 증권사 리포트 텍스트 분석[Industry Monitoring Using AI Algorithm: Analyst Reports Text Mining]. BOK이슈노트, No.2023-5. [pdf]


• Shin, SJ., & Seo, B., (2022). 빅데이터를 이용한 실시간 민간소비 추정[Real-Time Private Consumption Prediction Using Big Data]. BOK Working Paper, No.2022-16. [pdf]


• Seo, B.*, Lee, Y., & Cho, H., (2022). Machine-Learning-Based News Sentiment Index (NSI) of Korea. BOK Working Paper, No.2022-15. [pdf]


• 서범석, (2022). 뉴스 텍스트를 이용한 경기 예측: 경제부문별 텍스트 지표의 작성과 활용[Economic Forecasting Using News Texts: The Compilation and Usage of Text-based Economic Field Indicators]. BOK이슈노트, No.2022-18. [pdf]
  
  


• 서범석, 이영환, 조형배, (2022). 기계학습을 이용한 뉴스심리지수(NSI)의 작성과 활용[Comilation and Usage of News Sentiment Index (NSI) Using Machine Learning]. 국민계정리뷰, 2022(1), 68-90. [pdf]


• Seo, B.*, (2021). Interpretable Statistical Learning: From Hidden Markov Models to Neural Networks (Ph.D. Dissertation). Penn State University.

In Progress

• Chang, Y. -L., & Seo, B., Using Explainable AI to Explore the Key Features in the Future of Work, Workers, and Workplace.


• Seo, B., Explainable Neural Mixture Modeling For Clustering.


• Seo, B. & Yoo, J., Collective Expert Intelligence : Automated Algorithms for Industry Analysis Using AI and Language Modeling

MEDIA COVERAGE

News Coverage

• 한은 경제모형실, 빅데이터로 하반기 분석해보니 "소비 개선·물가 둔화" (아주경제, 2024.08.25)
• 한국은행, 단기 경제 전망 모형 '프리즘Now' 개발 (연합뉴스, 2023.12.19)
• AI 텍스트 분석, 숫자 넘어선 예측 가능 (연합인포맥스, 2023.02.16)
• 챗 GPT 등 자연어 기술로 경제 분석 혁신 될 것 (금융경제신문, 2023.02.16)
  ...
• (SBS 모닝와이드 - 친절한 경제, 2023.02.16)
  
  

Magazine

• 설명가능한 인공지능? 한은소식, 2021(9), 36-39.
• 텍스트마이닝으로 분석한 한은소식. 한은소식, 2020(5), 20-21.
  
  


• 머신러닝, 유학, 그리고 무언가를 배운다는 것. 한은소식, 2020(3), 11-13.

PROJECTS & PATENTS

• (co-I) 뱅크런 징후의 조기 포착을 위한 경보체계 구축 [Early Warning System Construction for Detecting a Bank Run], 2024, Korean Deposit Insurance Corporation
• (co-I) 비정형 금융경제 데이터의 정량화에 대한 연구 [Using Noise Filtering and Sufficient Dimension Reduction Method on Unstructured Economic Data], 2022, Bank of Korea
• (co-I) 빅데이터를 이용한 실시간 민간소비 추정 [Real-Time Private Consumption Prediction Using Big Data], 2021, Bank of Korea
• (Published Pattent) 뉴스 텍스트 기반 경제지표 작성장치, 텍스트 지표를 이용한 경기 예측 방법 및 시스템 [News Text-based Economic Indicator Calculation Device, Economy Prediction Method and System Using Text Indicators], (10-2022-0059258)

TEACHING

(Statistics & Engineering)

• Practicum in Statistical Analysis (UG in Statistics at SMU)
• Statistical Methods I (MS in Statistics at SMU)
• Machine Learning for Finance and Economics (MA in Statistics at Korea Univ.)
• Introduction to Probability and Stochastic Processes (UG in Engineering at PSU)
• Elementary Statistics (UG in Statistics at PSU)

(Economics & Business)

• Big Data Analytics in Macroeconomics (MA in Economics at SKKU)
• Topics in Data Information (MA in Grad School of Policy Studies at Korea Univ.)

EXPERIENCE

Committees

Bank of Korea

Penn State University

CODES

pip install tspoon

import mixturelinearmodel
from mixturelinearmodel import MixtureLinearModel
from utils import plot_mosaic, plot_ci, explainable_tree, explainable_condition, explainable_dim, highest_explainable_dim, plot_id_1d, plot_id_2d, plot_id_3d

install.packages("devtools")
devtools::install_github("seo-beomseok/HDclustVS")
library(HDclustVS)
?HDclustVS

install.packages("OTclust")
library(OTclust)
?OTclust

SEO Keywords: 서범석 한국은행 통계 머신러닝 경제