Developed a machine learning pipeline to classify the severity of road accidents using a multi-class target variable (Accident_severity) and 31 input features.
Applied end-to-end data science workflow including data cleaning, preprocessing (label encoding, scaling), exploratory data analysis (EDA), and model training.
Implemented Logistic Regression and Decision Tree Classifier using Scikit-learn.
Evaluated model performance using F1 Score, Confusion Matrix, and Classification Report to handle class imbalance and ensure reliable predictions.
Tools & Technologies: Python, Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn