Skip to content
View ahartshorn416's full-sized avatar

Block or report ahartshorn416

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
ahartshorn416/README.md

Hi! 👋 I'm Alison Hartshorn

💫 About Me:

I'm a Data Science Master's candidate (expected June 2026) with hands-on experience building end-to-end analytics pipelines at scale — including a 16M+ row machine learning project using U.S. Census data. I specialize in predictive modeling, fairness-aware ML, and translating complex findings into actionable insights for business and policy stakeholders. My background in legal operations and academic administration gives me an edge in data governance, compliance, and communicating with non-technical audiences.

🔭 Featured Projects:

  • 🏠 Rent Burden Prediction — Fairness & ML analysis on 16M+ ACS PUMS household records (Logistic Regression, Random Forest, Gradient Boosting); equity analysis across race, sex, and geography for HUD policy context
  • 📉 Customer Attrition Prediction — XGBoost churn model in R achieving 96% accuracy & AUC 0.99; SHAP values used to surface top business drivers
  • 🏦 Home Loan Approval Prediction — ML pipeline on 4.25M real HMDA 2023 mortgage applications; XGBoost ROC-AUC 0.9932, 96.3% accuracy across 121 features
  • 📊 Marketing Campaign Effectiveness — End-to-end ROI analysis for Nike Inc. using real Google Trends (pytrends API) + SEC EDGAR 10-K filings; ROAS modeling, lag correlation, and 6-panel dashboard in Python
  • 🏗️ Construction Project Management Dashboard — PostgreSQL analytics + interactive Tableau dashboard with KPI cards, risk scores, and delay trend analysis
  • ☕ Café Sales Data Cleaning & Analysis — SQL pipeline to clean and standardize transaction records; findings visualized in Tableau & Power BI

🌐 Socials:

LinkedIn Tableau

💻 Tech Stack:

Python R MySQL PostgreSQL NumPy Pandas Matplotlib scikit-learn XGBoost PyTorch Tableau Power BI SciPy

📊 GitHub Stats:



Pinned Loading

  1. predicting-rent-burden predicting-rent-burden Public

    ML models predicting U.S. household rent burden using 16M+ ACS survey records — includes fairness analysis across race, sex & geography to inform housing policy.

    Python

  2. home_loan_approval_prediction home_loan_approval_prediction Public

    Predicts U.S. home loan approvals using 4.25M real HMDA 2023 applications — XGBoost, Random Forest, Logistic Regression, ROC-AUC 0.9932

    Python

  3. marketing-effectiveness-analysis marketing-effectiveness-analysis Public

    Analyzing marketing campaign ROI for Nike using Google Trends, SEC EDGAR financials, and lag correlation — built in Python with pandas, matplotlib, and pytrends.

    Python

  4. customer-attrition-prediction customer-attrition-prediction Public

    End-to-end customer churn prediction in R comparing Logistic Regression, Random Forest & XGBoost (AUC 0.99) — with SHAP values for interpretability.

    R

  5. construction-pm-analytics construction-pm-analytics Public

    PostgreSQL analysis of construction project task data to surface overdue work, bottlenecks & risk scores — visualized in an interactive Tableau dashboard.

  6. cafe-sales-data-cleaning-with-sql cafe-sales-data-cleaning-with-sql Public

    SQL data cleaning pipeline + Tableau & Power BI dashboards analyzing café sales revenue, product performance, and payment trends.