I am passionate about transforming raw data into actionable business insights. I specialize in building predictive models, data visualization, and end-to-end analytics pipelines.
-
E-Commerce Customer Churn Prediction: Built an end-to-end Machine Learning pipeline using XGBoost & SMOTE, and integrated the predictions into an interactive Power BI dashboard to help businesses reduce customer attrition.
-
Credit Risk Prediction (ANN): A Deep Learning approach to predict loan default risk using PyTorch. Utilized an Artificial Neural Network (ANN) to classify borrowers based on financial history, achieving ~92% accuracy.
-
Fraud Detection Analysis: Analyzed financial transactions to detect fraud using Random Forest, XGBoost, LightGBM, and Logistic Regression, focusing on imbalanced dataset handling.
-
Temperature Prediction (Time-Series): A comparative time-series analysis predicting daily temperature using Holt-Winters, Facebook Prophet, and XGBoost with Lag Features.
- Programming: Python, SQL, JavaScript, HTML
- Data Science & ML: Pandas, NumPy, Scikit-Learn, XGBoost, LightGBM, PyTorch
- Data Visualization: Power BI, Matplotlib, Seaborn
- Tools & Workflow: Git, GitHub, Jupyter Notebooks
- LinkedIn: Nicolaus Prima
- Email: nicolausprimaa@gmail.com