Zehui Gao Email & Phone Number

Marina, California, United States

Marina, California, United States

Virginia Beach, Virginia, United States

• Introduced K-means clustering algorithm to the marketing penetration analysis and successfully identify 4 distinct patterns within 41 branches which help marketing team optimize 2022 AOP plan.• Developed feature selection methods based on Boruta algorithm and Pearson correlation for RRFM matrix. Constructed RRFM matrix for detecting most valuable zip code areas.• Created an automated ETL pipeline in python that can pull, clean, transform and aggregate data from Google BigQuery database. Shorten the turnaround time from 3 days to 1 hour.• Built a 3-page Dashboard on Google Data Studio for Paid Search Channel to visualize the topline performance MoM, and realize daily performance tracking by branch.

San Diego County, California, United States

• Project: GPI Enhancement and Maintenance Based on Customer Satisfaction Data. Analyzed the sample data collected by the Client (validation/modification/visualization /exploratory). Creatively proposed a linear regression model to extract insights and done proof of concept (POC) 2 weeks in advance. Modified data-cleaning pipeline and cleaned the big data.• Project: Credit Analysis - Predict Probability of 90 Days Past Due DelinquencyBuilt three models based on Logistic Regression / Random Forest / XGBoost and evaluated the predictive performance.Optimized the model based on XGBoost algorithm to achieve AUC of 0.866.

San Diego, California, United States

Taught Stat 119A: undergraduate recitation courses on topics including, but not limited to, statistical theory, probability, linear regression, hypothesis testing.Hosted office hours and instructional sessions to provide personalized teaching and additional learning environments.