Résumé

• Developed a predictive model to forecast daily electricity demand using data from the U.S. Energy Information Administration (EIA) and NOAA weather stations
• Collected and preprocessed datasets, including electricity demand and time-series weather variables
• Performed data cleaning, handling missing values, and merging multiple datasets

Software: Python / Pandas / Matplotlib

• Integrated FRA railroad accident data with USDA Rural–Urban Continuum Codes to analyze accident risk across urban and rural environments
• Performed data cleaning, exploratory data analysis, and feature engineering to improve dataset quality
• Prepared structured train/test datasets for predictive modeling and machine learning analysis

Software: Python / Pandas / Jupyter Notebook

• Developed a data acquisition and preprocessing pipeline to integrate NASA Exoplanet Archive and EU Exoplanet datasets using TAP-based queries
• Built data cleaning and merging workflows in Pandas to construct a unified multimodal exoplanet dataset for scientific analysis
• Organized and processed large-scale astronomical datasets in Jupyter Notebook to support exploratory analysis and machine learning workflows

• Developed a data analysis pipeline to evaluate water potability and identify models with low prediction error
• Performed data cleaning, feature analysis, and error evaluation to improve classification accuracy
• Created visual statistical summaries (distributions, correlations, and trends) to enhance interpretability and user understanding