Portfolio - Foster Oppong

Hello, I am

Foster Oppong

Data Scientist | ML Engineer

Download CV

About Me.

Hi, I'm a Data Scientist using data to predict trends to make impactful business decisions.

1

Machine Learning Project

Skills.

Python

Pandas

Numpy

Sci-kit learn

Machine Learning

SQL

Projects.

My Experience.

Data Analyst - Wingate University Advancement Services, Wingate, NC

January 2024 - Present

Conducted statistical analyses on 10K+ alumni records, optimizing fundraising strategies and improving database accuracy.
Developed predictive models using Python and SQL to identify top alumni donors, leading to a 20% increase in fundraising event pledges
Automated data imports and cleaned large datasets, improving data accuracy and reliability by 30%.

Awards & Certificate.

Python: Beginner to Advanced For Data Professionals

SQL for Data Science

SQL Beginner to Advanced For Data Professionals

Math and Statistics For AI, Data Science

Master Machine Learning for Data Science & AI: Beginner to Advanced

Contact.

Feel free to get in touch with me. I am always open to discussing new projects, creative ideas or opportunities to be part of your visions.

Download CV

[email protected]

Call Me

+1 9802100083

https://linkedin.com/in/foster-oppong

GitHub

https://github.com/data-detectiv

Machine Learning Project

Healthcare Insurance Prediction

Domain/Function: Insurance

Helped an Insurance company to target customers who are likely to subscribe to an insurance premium. This project helped the company to reduce cost on targeting the wrong customers.

View Project
on GitHUB

This is the UI for inputing the data.

This is the second one.

Project Details

• Developed a machine learning model to predict health insurance premiums with 98% accuracy, improved pricing precision.
• Processed and cleaned 50,000+ data points, performed feature selection, and engineered 10+ key features to enhance model performance.
• Optimized health insurance premium prediction model by fine-tuning XGBoost, Linear Regression, Lasso, and Ridge regressions via RandomizedSearchCV, decreasing prediction error by 30% and improving premium accuracy.
• Helped insurers optimize premium pricing strategies, potentially reducing underpricing risks by 25% based on model insights.