Hours:
20 hours (5 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa - Ground Floor

To register to the course, click here

Short Abstract:

Recommender Systems have transformed user experiences across industries such as e-commerce, media streaming, and healthcare [1]. This course, tailored for PhD students, offers a comprehensive introduction to RS algorithms, covering foundational concepts like collaborative filtering and matrix factorization, emphasizing how Recommender Systems use user data to generate personalized recommendations. The course will delve into the synergy between AI and Recommender Systems,
teaching machine learning techniques for Recommendation and introducing deep learning for complex user preferences and item relationships [2].
The course will also showcase the power of Recommender Systems in specific fields, demonstrating their practical applications and impact.
In Telecommunications [3,4,5], Recommender Systems can be employed to personalize network resource allocation by analyzing user profiles and traffic patterns, ensuring efficient and userspecific network management.
In Biosciences [6,7,8], Recommender Systems can assist in drug discovery by recommending promising drug candidates based on complex data analyses, or be implemented for personalized medicine, where targeted therapies are suggested according to individual patient profiles, leading to more effective treatments.
In the field of Automation Engineering [9,10,11], Recommender Systems are utilized to analyze sensor data, predicting component degradation and recommending maintenance schedules to ensure optimal performance and longevity of devices.
By the end of the course, students will gain a thorough understanding of Recommender Systems principles and their applications in diverse research domains. They will be equipped with the practical skills and tools necessary to implement and evaluate their own recommender systems using Python libraries such as NumPy, Pandas, Scikit-learn, and TensorFlow/PyTorch for deep learning applications.

Course Contents in brief:

1. Foundational Concepts:
   1. Define Recommender Systems and their impact across various industries.
   2. Explore core algorithms like collaborative filtering, Markov chains, and hybrid approaches.
   3. Discuss scalability challenges and optimization techniques for large-scale implementations.
   4. Analyze case studies of successful implementations from diverse domains.
2. Machine Learning and Neural Networks for Recommendations:
   1. Learn key AI techniques used in Recommendation: matrix factorization, neighborhood methods, and an introduction to deep learning for recommendations.
   2. Understand the role of neural networks in building advanced recommender systems that capture complex user preferences and item relationships.
   3. Explore various neural network architectures commonly used for recommendations, such as Recurrent Neural Networks and Transformers.
3. Telecommunications:
   1. Personalization of network resource allocation based on user profiles and traffic patterns.
   2. Optimization of content delivery with recommendation of appropriate caching strategies.
   3. Development of recommender systems for suggesting network service plans tailored to individual customer needs.
4. Biosciences:
   1. Analyze vast patient data and molecular structures using Recommender Systems for drug discovery, recommending promising drug candidates.
   2. Implement systems for personalized medicine by recommending targeted therapies based on individual patient profiles.
   3. Utilize Recommender Systems to recommend targeted interventions for disease prevention and management.
5. Automation Engineering:
   1. Utilize Recommender Systems to analyze sensor data and recommend maintenance schedules based on predicted component degradation.
   2. Implement Recommender Systems to recommend optimal component sourcing and production strategies based on real-time demand and inventory data.
   3. Design Recommender Systems to recommend product configurations tailored to individual customer needs and specifications.

Schedule:

1. 5/5/2025 – 14:30-18:30
2. 6/5/2025 – 14:30-18:30
3. 7/5/2025 – 14:30-18:30
4. 8/5/2025 – 14:30-18:30
5. 9/5/2025 – 14:30-18:30