For my Semester 4 midterm project, I conducted a comprehensive heart failure prediction analysis using Jupyter Notebook. Through Exploratory Data Analysis (EDA), I visualized key trends, handled missing values and outliers, and engineered new features such as age groupings and BMI to enrich the dataset. I categorized variables, split data into training and testing subsets, and built predictive models including logistic regression, decision trees and random forests. These models were rigorously evaluated using accuracy, precision, recall and F1-score. I identified critical predictors like ejection fraction and serum creatinine, offering clinical insight into patient outcomes and proposing actionable recommendations to support healthcare decision-making. This project sharpened my ability to translate raw data into meaningful, health-oriented solutions through transparent, reproducible workflows.
Note: other's code or resources used are referenced.