Recommender Systems

This course on recommender systems delves into the fundamental concepts that shape modern digital interactions and decision-making. By analyzing how recommendation algorithms are used in platforms like streaming services, ecommerce, and social media, students will understand the mechanisms that influence user engagement and behavior. The course explores the development and application of collaborative filtering, content-based recommender, and hybrid models to predict and recommend personalized content. Additionally, it emphasizes practical methodologies for conducting offline experimentation and measuring the effectiveness of these systems, ensuring that students gain a comprehensive view of the evaluation process. The course will also take a critical look at the challenges and societal issues inherent in creating and maintaining recommender systems. Topics include data sparsity, algorithmic bias, and the need for transparency in model decisions. Students will work on practical projects using popular tools and frameworks, gaining experience in data preprocessing, model training, and model evaluation. The course further emphasizes the critical analysis of recommendation outcomes. students will learn to analyze the recommendation results, discussing potential ethical and societal implications. This includes reflecting on how recommender systems can influence user behavior and societal norms, highlighting issues such as filter bubbles and content diversity. By the end of the course, participants will have developed the skills to design, implement, evaluate, and analyze recommender systems.