Skip to content
View SOOBEENKIM's full-sized avatar

Organizations

@finxlab

Block or report SOOBEENKIM

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
SOOBEENKIM/README.md

Soobeen Kim — Financial Engineering and Machine Learning, Hanyang University

Integrated M.S.–Ph.D. Student · Industrial Engineering · Hanyang University
Financial Engineering Lab · Master's stage, second semester

FINX Lab Lab GitHub LinkedIn ORCID: 0009-0004-7688-1921

Current Focus   /   Achievements   /   Research   /   Project Collection


I am a second-semester student in the integrated M.S.–Ph.D. program in Industrial Engineering at Hanyang University, affiliated with FINX Lab. I earned my bachelor’s degree in Industrial Engineering at Hanyang and joined FINX Lab as an undergraduate researcher before continuing my graduate studies in the same lab.

My research interests lie in financial machine learning, time series analysis, data generation, and LLM-based agents.

Current Focus

Research / collaboration Focus / role Status
Stablecoins Stablecoin research Ongoing research
Data Generation Data generation research Ongoing research
Neural ELS Hedging Neural network–based optimal control for ELS hedging Ongoing collaboration
WorldQuant BRAIN Research Consultant Current role

Competitions & Achievements

Competition Result / status
KDD Competition 🏆 1st Place
Mirae Asset AI Festival Participated · Results pending

Selected Research Projects

Enhancing Commodity Factor Strategies with Deep Learning

Applied LSTM and Transformer models to dynamically rank 19 commodity futures based on basis-momentum (BMOM), a hybrid factor combining momentum and term structure. Code

Volatility Prediction for Bitcoin using LSTM-GARCH Hybrid Model

Integrated LSTM and GARCH models to jointly predict return and volatility of Bitcoin. Demonstrated improved forecasting accuracy under high volatility regimes. Code

Impact of Audit Reform on MD&A Tone and Performance Deviations

Conducted a text mining analysis to quantify tone shifts in MD&A before and after Korea’s New External Audit Act. Used Difference-in-Differences methodology to show significant increases in abnormal tone post-reform, suggesting improved strategic language use and enhanced investor transparency.

Deep Learning-Based Malware Generation and Classification

Developed an LSTM-based sequence autoencoder to extract latent vectors from malicious Python code across 11 behavioral types. Trained a conditional GAN (CGAN) to generate malware-type-specific latent vectors, which were decoded back to source code using autoencoder decoders. Evaluated classification accuracy, reconstruction error, and semantic consistency of generated samples. Code

Project Collection

A collection of projects spanning quantitative finance, data science, experimental research, software, and interdisciplinary design. Expand each category to explore the projects.

Quantitative Finance & Data Science · 5 projects

Financial Engineering Projects

Modern Portfolio Theory, CAPM and Factor Model, Closed-form BlackScholesMerton pricing model, Binomial tree Code / Code

Time Series Forecasting: ARMA for Air Passenger Demand

Forecasted monthly air passenger volume using ARMA modeling. Evaluated forecasting accuracy and seasonality components.

P2P Loan Approval Prediction (SVM Model)

Built a classification model using SVM to predict loan approval decisions based on borrower-level financial attributes from real-world P2P lending data. Code

Used Car Price Prediction (Ensemble Models)

Applied Random Forest and AdaBoost regression to predict used car market prices. Preprocessed real-world listings and evaluated performance using RMSE. Code

Q-Learning in Maze Navigation (Numerical Analysis Project)

Explored the effect of discount factor (gamma) on Q-learning convergence when solving a maze. Analyzed exploration-exploitation tradeoffs using MATLAB; found minimal difference in convergence time on small maps. Code

Experimental & Interdisciplinary Research · 5 projects

Experimental Design Using MINITAB: Cold-Temperature Adhesion Study

Applied a 2⁷⁻³ fractional factorial DOE to analyze effects of phosphoric acid deoxidizer, neutralizer concentration/time, and adhesion promoters on aluminum bonding strength at –65°F. Identified significant factor interactions impacting low-temp peel strength.

Steam R&E : Microbiome Analysis in Fermented Foods

Investigated microbial diversity changes in Korean fermented foods (cheonggukjang, jeotgal, kimchi) across fermentation periods using 16S rRNA sequencing via Ion Torrent and Ion Reporter pipelines. Results showed pH/salinity-driven shifts in microbiome structures. Code

Frontier Chemistry: Lithium Extraction from Seawater

Designed ion-exchange resins using plant-based cellulose to selectively absorb lithium ions from seawater. Compared Li absorption efficiency across different functional groups and bead formation conditions.

STEAM R&E : Gill-Inspired Fine Dust Capture Device

Designed a cross-step filtration system modeled after fish gill structures to capture fine dust in vehicle exhaust. Used SolidWorks and Flow Design to simulate fluid flow, and built a detachable .STL-based prototype optimized for safety and energy efficiency. Code

Acoustic Phase Control for Directional Speakers (Physics)

Developed a directional sound system using acoustic phase control. Applied MATLAB to optimize speaker array configurations and control signal origin with function generator input.

Software, Business & Project Design · 4 projects

Impact Business Project: Coffee Grounds Upcycling

Developed a solution to encourage coffee shops to participate in spent grounds collection through a subscription model and eco-friendly platform. The initiative aimed to raise public awareness of coffee waste as a recyclable resource.

OOP Projects – Game Development in Java

Built generalized rock-paper-scissors game and a “catch-the-mouse” strategy game in Java as part of team assignments in object-oriented programming.

Digital Literacy Startup Proposal using ChatGPT

Proposed a digital literacy education service powered by ChatGPT. Designed curriculum frameworks targeting students and non-technical users. Code

Project Management Engineering – Seminar Presentation

Delivered a presentation on “Make Megaprojects More Modular”, discussing modularization benefits in large-scale engineering projects.

Technical Toolkit

Area Technologies & methods
Programming Python · R · SQL · C · C++ · Java · MATLAB
Machine Learning PyTorch · TensorFlow · Keras · Scikit-learn · XGBoost · LSTM · GAN
Data & Statistical Analysis Pandas · NumPy · STATA · MINITAB · GARCH · NLP · Time Series Forecasting
Development Git · GitHub · Jupyter · VS Code · Anaconda · Docker
Broader Interests & Background

Investment Science · Financial Engineering · Forecasting · Machine Learning · Data Mining · Optimization · Operations Research · Cybersecurity · Data Analytics · ADSP

Education

  • Hanyang University — Integrated M.S.–Ph.D. in Industrial Engineering
    Currently enrolled · Master's stage
  • Hanyang University — Bachelor's degree in Industrial Engineering
    Graduated

Financial Engineering · Machine Learning · Time Series · Data Generation

Popular repositories Loading

  1. LSTM-GARCH LSTM-GARCH Public

    Jupyter Notebook 1

  2. P2P_Loan P2P_Loan Public

    P2P Classification

    Jupyter Notebook 1

  3. UsedCar_Adaboost UsedCar_Adaboost Public

    UsedCar Prediction

    Jupyter Notebook 1

  4. FinancialEngineering1 FinancialEngineering1 Public

    using stock data

    Jupyter Notebook 1

  5. FinancialEngineering2 FinancialEngineering2 Public

    BlackScholesMerten

    Jupyter Notebook 1

  6. Q-Learning Q-Learning Public

    maze problem

    MATLAB 1