Skip to content
View dennis20413's full-sized avatar
1 follower · 1 following

Block or report dennis20413

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 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
dennis20413/README.md

Hi, I'm Dennis Chen 陳雲皓 👋

Data Science student (GPA 3.9x) who turns messy, real-world data into models people can actually use and understand.

I build end-to-end: data pipelines, feature engineering, machine learning, and clear evaluation — with reproducible, tested code. My work spans applied ML, NLP, analytics, and quantitative finance.

🧰 Tech Stack

Python pandas NumPy scikit-learn SQL Streamlit

  • Languages: Python, SQL, VBA
  • ML & Stats: Classification, regression, NLP, feature engineering, cross-validation, model evaluation & interpretability
  • Analytics & Data Eng: SQL (window functions, CTEs, cohorts), EDA & visualization, REST API ingestion, large-scale data cleaning
  • Apps: Streamlit (interactive, deployable ML apps)

📌 Featured Projects

🤖 Machine Learning & Data Science

Project What it demonstrates
LoL Win Predictor — Live App End-to-end & deployable: a trained model wrapped in an interactive Streamlit web app
Esports Win Predictor (ML + Riot API) ML classifier with feature engineering, cross-validation, and a real Riot API data pipeline
NLP Sentiment Analysis Text preprocessing (negation handling), TF-IDF, interpretable classification
Customer Insights EDA Data-quality auditing, segment profiling, correlation analysis & visualization

🗄️ Data & Analytics

Project What it demonstrates
SQL Sales Analytics Window functions, CTEs, cohort retention, RFM segmentation
Forensic Data Cleaning Pipeline Multi-phase cleaning & reconciliation of a 50k-row dirty dataset

📈 Quantitative Finance

Project What it demonstrates
Multi-Factor Stock Model Factor construction, Information Coefficient evaluation, quantile portfolios
GARCH Volatility & VaR Time-series volatility modeling, forecasting, Value-at-Risk
Pairs Trading — Stat Arb Cointegration testing, z-score signals, market-neutral backtest
Portfolio Optimization Mean-variance, min-variance & risk-parity allocation
Quant Backtesting Framework Strategy backtesting with realistic costs and no look-ahead bias
Piecewise Regression — Equity Analysis Statistical trend decomposition, validated against a reference tool

Every project includes unit tests and an honest methodology write-up — limitations stated, not hidden.

📫 Reach me

LinkedIn   📧 dennis07250725@gmail.com

Currently a junior building toward data science / machine learning roles.

Pinned Loading

  1. factor-model factor-model Public

    Cross-sectional multi-factor stock selection — momentum/volatility/reversal factors, Information Coefficient evaluation, quantile portfolios, and composite alpha.

    Python

  2. eda-customer-insights eda-customer-insights Public

    Exploratory data analysis of customer data — data-quality auditing, segment profiling, correlation analysis, churn drivers, and presentation-ready visualizations.

    Python

  3. esports-win-predictor-LoL esports-win-predictor-LoL Public

    Machine-learning classifier predicting League of Legends match outcomes from mid-game snapshots — feature engineering, cross-validated training, interpretable feature importances, and a real Riot A…

    Python

  4. esports-winrate-app esports-winrate-app Public

    Interactive Streamlit web app predicting League of Legends win probability in real time from the live game state — an end-to-end, deployable ML demo.

    Python

  5. nlp-sentiment-analysis nlp-sentiment-analysis Public

    End-to-end NLP sentiment classifier — text preprocessing with negation handling, TF-IDF features, interpretable logistic regression, and honest evaluation.

    Python

  6. sql-sales-analytics sql-sales-analytics Public

    Analytical SQL portfolio on an e-commerce dataset — joins, window functions (RANK/LAG/NTILE), CTEs, cohort retention, and RFM segmentation.

    Python