Building practical solutions
from data, systems, and process experience.

Built a supervised machine learning model to predict whether a telecom customer would choose the Ultra mobile plan based on monthly usage behavior including calls, minutes, messages, and internet traffic. The project demonstrates classification modeling, structured evaluation, and practical business-focused decision support.

• Prepared and analyzed structured customer usage data using Python, Pandas, and Jupyter Notebook
• Split data into training, validation, and test sets for proper model evaluation
• Compared Decision Tree, Random Forest, and Logistic Regression models using scikit-learn
• Tuned model parameters and selected the strongest performer
• Confirmed model accuracy exceeded the project threshold of 0.75
• Demonstrated an end-to-end ML workflow including preprocessing, model comparison, and performance validation

Explored entertainment data to uncover trends in highly rated films, actor presence, and release-period patterns while practicing structured data cleaning and reusable analysis logic.

• Cleaned and standardized raw data using Pandas
• Built reusable functions for decade filtering, score categorization, and actor extraction
• Performed exploratory analysis to identify top-rated content patterns
• Documented work clearly in Jupyter Notebook for readability and reproducibility
• Strengthened debugging, data storytelling, and analytical reasoning skills