Skip to content
View lucytheboss's full-sized avatar

Highlights

  • Pro

Organizations

@LUCAUS2025

Block or report lucytheboss

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
lucytheboss/README.md

Hi, I'm Jiwoo (Lucy) Roh 👋

I analyze user behavior to understand why people hesitate, drop off, or fail to act in digital systems, using behavioral data and HCI-informed reasoning.


Who Am I

Behavioral UX Research · Human-Computer Interaction (HCI)

  • 🏫 Undergraduate @ Chung-Ang University
  • 📚 Majoring in Psychology, English Literature, and Technology-Art

I focus on explaining why users behave the way they do, especially in moments of hesitation, confusion, or disengagement.

Rather than optimizing interfaces or models in isolation, my work emphasizes behavioral interpretation — identifying cognitive, contextual, and systemic factors that shape user action (or inaction).


What I Work On

  • 🧠 Behavioral Science
    Cognitive psychology, user decision-making, behavioral hypothesis building

  • 📊 Behavioral Data Analysis
    Interaction logs, behavioral metrics, exploratory analysis, residual-based diagnostics

  • 🎨 UX & Interaction Research
    UX research methods, interaction flow analysis, prototyping for behavioral inquiry


🔍 Selected Projects

Project Focus Key Outcome
🎧 Music Discovery Gaps: Behavioral Signals Behind Popularity Failures Behavioral data science · discovery systems Built a residual-based popularity model (R² = 0.20) to analyze when recommendation predictions fail, revealing 20% unexpected outcomes, genre-level exposure bias, and underexposed “hidden gem” tracks. Delivered a full pipeline (Python modeling → report → Tableau dashboard) to diagnose discovery inefficiencies.
🎬 Behavioral Recommender Systems: Understanding User Choice with MovieLens Recommendation behavior Modeled user–item interaction patterns and collaborative filtering signals to analyze how familiarity, preference history, and exposure shape recommendation outcomes and behavioral choice.
🎪 Why Users Stopped: Behavioral Analysis of a University Festival Website UX drop-off analysis Identified points of hesitation and abandonment through real interaction logs, showing how unclear next-action cues disrupted behavioral continuation and task completion.
🎧 How Audio Platforms Shape Listener Expectations Human interaction Explored how audio features & artist momentum influence perceived popularity through interactive behavioral analysis and signal interpretation.

🛠 Tools

Analysis & Modeling

Python Pandas scikit-learn SQL

Visualization, Apps & Reporting

Tableau Streamlit Matplotlib Jupyter

UX & Research

Figma Interaction Logs

Workflow

Git


What I’m Looking For

I’m interested in roles and research environments where understanding
why users fail to act matters more than maximizing model accuracy.

I’m particularly drawn to UX Research, Behavioral Data, HCI, and Human–AI Interaction problems that require explaining behavior — not just predicting outcomes.

Pinned Loading

  1. creative-coding creative-coding Public

    Processing

  2. festival-behavioral-analytics festival-behavioral-analytics Public

    Chung-Ang University 2025 Spring festival website

    Jupyter Notebook

  3. Hit-Song-Predictor Hit-Song-Predictor Public

    Apple Music Data Analysis

    Jupyter Notebook

  4. Sonic-Analytics-Hit-Predictor Sonic-Analytics-Hit-Predictor Public

    Search for songs and see their audio features analyzed!

    Jupyter Notebook