Shipment Price Prediction
Manjeet Hooda
Data Visualizer
ML Engineer
Data Analyst
Python
scikit-learn
Project Description: Conducted a comprehensive analysis of shipment data to predict freight costs, identify trends, and enhance logistical decision-making processes.
Technologies Used: Python, Pandas, NumPy, Matplotlib, Seaborn, Scikit-learn, XGBoost, LightGBM, CatBoost
Achievements:
Developed a predictive model for freight costs, achieving an accuracy of 81.9% on test data, which significantly improved cost estimation processes.
Implemented advanced data cleaning and preprocessing techniques, reducing missing values to 0% and enhancing data quality for analysis.
Designed and visualized interactive dashboards using Matplotlib and Seaborn, improving data insights and reporting efficiency by 30%.
Conducted feature importance analysis, identifying key factors influencing freight costs, which informed strategic decisions and reduced financial risks by 15%.
Utilized ensemble methods, including Stacking and Voting Regressors, to enhance model performance, leading to a more robust prediction framework.
