Open to opportunities

Keho Olamide Stella

@kehoolamidestella

Message

Data Analyst turning complex data into actionable insights with SQL, Power BI, Python.

Nigeria

Message

What I'm looking for

I’m looking to grow in an analytics-focused role where I can build dashboards, write SQL reports, and apply machine learning/NLP to deliver actionable insights. I’m ready to work hard, learn fast, and support data-driven decisions.

I’m an economics graduate and Data & Financial Analysis Projects Analyst who turns large, messy datasets into clear reporting and decision-ready insights. I build financial models and dashboards, write SQL queries to extract and aggregate data, and deliver analytical reports that stakeholders can act on.

In my projects, I combine analytics with machine learning and NLP—using techniques like Logistic Regression with TF-IDF for fake news detection, and predictive modeling for insurance premium and employee burnout risk. I also deploy an interactive model through a Streamlit app, and I’m driven by analytical rigor, clear communication of findings, and continuous improvement.

Experience

Work history, roles, and key accomplishments

Employee Burnout Predictor

Neuro Well Analytics

Built a machine learning model to predict employee burnout risk based on work conditions, stress levels, and mental fatigue. Improved prediction accuracy using EDA, SMOTE, and feature engineering, then deployed the final model in a Streamlit app.

Machine Learning EDA SMOTE Feature Engineering Classification Python Data Cleaning Streamlit

Insurance Premium Prediction

Secure Life Insurance Co.

Analyzed 200K+ customer records to predict insurance premium amounts using feature engineering and log transformation. Trained a HistGradientBoostingRegressor and documented results for reporting and presentation.

Regression Modeling Log Transformation Feature Engineering HistGradientBoostingRegressor EDA data preprocessing Python Data Reporting Visualization

Fake News NLP Classifier

Truth Lens Institute

Built a fake news classifier using Logistic Regression with TF-IDF features. Performed text cleaning, lemmatization, topic modeling, and POS tagging to identify linguistic patterns distinguishing fake vs. real news.

NLP Logistic Regression TF IDF Topic modeling POS Tagging Lemmatization Supervised Learning Model Evaluation