Data

Global Stability & Risk Forecasting (GDELT) featured image

Global Stability & Risk Forecasting (GDELT)

Analyzed subsets of a 2PB global events dataset to identify risk trends and support stability forecasting, including anomaly detection and a validated Random Forest model.

TransLink: Transit Insights (ETL + Dashboards) featured image

TransLink: Transit Insights (ETL + Dashboards)

Built end-to-end ETL pipelines with PySpark/Spark SQL to ingest 23GB of logs into a Data Lake using Medallion Architecture, then delivered KPI dashboards in Tableau/Power BI.

Steam-200k Recommender System (Implicit ALS) featured image

Steam-200k Recommender System (Implicit ALS)

Built a Top-N recommender using implicit-feedback ALS on the Steam-200k dataset, learning player preference profiles from gameplay behavior and recommending similar games.