WNSData Analyst Intern
Jun. 2024 - Sep. 2024Kolkata, West Bengal, India- Developed a sophisticated graph-based clothing recommendation system using Python, NetworkX, and machine learning techniques, capable of processing and analyzing over 90,000 unique items with their complex metadata and relationships. - Implemented advanced algorithms for outfit generation and diversity enhancement, combining graph theory, cosine similarity, and custom heuristics to ensure varied and relevant recommendations. - Designed and built a scalable and modular codebase integrating various Python libraries (NumPy, Pandas, Scikit-learn) for efficient data manipulation, feature extraction, and similarity calculations, while implementing robust error handling for missing or inconsistent data. - Gained extensive experience in recommendation system development, including data preprocessing, graph database management, algorithm optimization, and evaluation using metrics such as precision, recall, and diversity scores. - Demonstrated proficiency in solving complex e-commerce and fashion tech challenges, balancing computational efficiency with recommendation quality, and creating a user-friendly interface to showcase the system's capabilities.