JOHN AGU MICHAEL

Data Scientist | Machine Learning | Deep Learning | Python, SQL

24 Chimana Street, Rukpokwu, Port Harcourt, Nigeria.

About

Data Scientist with hands-on experience in Python, SQL, and machine learning, specializing in data cleaning, exploratory data analysis, feature engineering, and data visualization. Skilled in developing and evaluating predictive models using Scikit-learn and gradient boosting techniques, with experience applying machine learning to real-world problems. Strong analytical and problem solving abilities with a focus on extracting meaningful insights, evaluating model performance, and supporting data driven decision making.

Work

Freelance

|

Data Analyst

→

Contract

Summary

Performed customer transaction data analysis to evaluate transaction volume per customer, detect high-value transactions, measure transfer frequency, and identify anomalous spikes in activity using analytical techniques for pattern recognition and decision support.

Highlights

Queried and extracted structured data using SQL for analytical processing and reporting.

Analyzed 10,000+ customer transaction records using SQL to identify high-value transactions and behavioral patterns.

Detected anomalous transaction spikes using statistical techniques, improving fraud detection insights.

Performed frequency and trend analysis to uncover customer activity patterns and support financial decision making.

Delivered insights that highlighted key risk indicators and transaction behaviors.

Freelance

|

Data Analyst

Port Harcourt, Rivers, Nigeria

→

Contract

Summary

Executed a freelance project analyzing sales and business performance data, gaining experience in tracking revenue trends and understanding customer behavior. Delivered actionable insights to support business decision-making and operational planning.

Highlights

Cleaned, validated, and standardized real-world datasets using Python (Pandas).

Performed exploratory data analysis (EDA) to identify trends, patterns, and anomalies.

Analyzed sales and revenue data to track monthly growth and customer behavior.

Built dashboards to visualize KPIs, top-performing products, and revenue trends.

Automated reporting workflows, reducing manual analysis time by 30%.

Communicated insights clearly to support business and operational decisions.

Skill Boost Africa

|

Data Analyst Intern

Port Harcourt, Nigeria

→

NYSC

Summary

Gained hands-on experience analyzing real-world datasets and applying data analysis techniques during an NYSC internship program.

Highlights

Worked with structured datasets (CSV, Excel, SQL tables) for analysis and reporting.

Applied data cleaning techniques: handling missing values, removing duplicates, and standardizing formats.

Created visual reports and dashboards using Tableau and Power BI.

Supported decision-making by summarizing insights in clear, non-technical language.

Projects

Cryptocurrency Price Prediction

→

Summary

Developed an end-to-end cryptocurrency forecasting project using historical Solana (SOL) market data, including data preprocessing, exploratory data analysis, feature engineering, and time-series analysis. Implemented and compared Naive, Historical Average, and Window Average baseline models against Random Forest and XGBoost regression models. Evaluated model performance using MAE and RMSE with time-based validation and visualization of actual vs. predicted prices. Analyzed cryptocurrency price and return distributions and generated multi-day forecasts to assess model performance and predictive capability.

Customer Churn Prediction & Risk Analysis

→

Summary

Designed and deployed a machine learning pipeline for predicting customer churn behavior. Conducted exploratory data analysis to identify key drivers of churn such as age, balance, and product usage. Engineered categorical and numerical features using pipeline-based preprocessing. Trained and optimized classification models with hyperparameter tuning. Delivered a model with strong recall performance to reduce missed churn predictions.

Loan Default Risk Prediction

→

Summary

Developed an end-to-end classification model to predict borrower loan default risk. Performed data cleaning, missing value treatment, and feature engineering to improve model performance. Applied one-hot encoding and feature scaling while preventing data leakage through proper train-test splitting. Trained and compared Logistic Regression, Random Forest, and Gradient Boosting models. Evaluated models using Accuracy, Precision, Recall, F1-score, ROC-AUC, and Confusion Matrix. Saved trained model for inference on unseen data

Education

Clifford University

Abia State, Nigeria
→
B.Sc. Computer Science

Skills

Soft Skills

Communication, Problem solving, Team Collaboration, Adaptability.

Programming Language

Python.

Technical Skills

Machine Learning, SQL Querying and Data Manipulation, Exploratory Data Analysis(EDA), Data Wrangling, Data Scraping, Time Series Forecasting, Model Evaluation & Validation.

Certificates

Data Science Diploma

AltSchool Africa

Languages

English

Native