Omid Shojaeian Zanjani

This project presents a comprehensive Python implementation of Non-negative Matrix Factorization (NMF), incorporating multiple initialization techniques and multiplicative update algorithms to ensure flexible and robust matrix factorization. Designed to simulate the NMF process, the implementation supports diverse use cases by enabling effective decomposition of non-negative data matrices into interpretable lower-dimensional factors.

Additionally, this project explores the application of NMF in building content-based recommender systems, leveraging the factorized components to identify latent features and similarities between items or users. By experimenting with different initialization strategies and update rules, users can gain insights into the model’s convergence behavior and its effectiveness in generating personalized recommendations. This makes the implementation a valuable tool for tasks such as data mining, pattern recognition, dimensionality reduction, and recommendation engines.

In 2023, I had the opportunity to participate in the Apple Developer Academy at Federico II University in the beautiful city of Naples, Italy. This experience marked the beginning of my professional career in technology. During this immersive program, I developed strong skills in app development, user experience design, and AI integration. The Academy provided me with hands-on experience working on real-world projects, collaborating with diverse teams, and delivering innovative digital solutions. This foundation has been instrumental in my growth as a developer and AI practitioner, enabling me to approach complex challenges with creativity and technical expertise.