MSE Master of Science in Engineering

The Swiss engineering master's degree

Chaque module vaut 3 ECTS. Vous sélectionnez 10 modules/30 ECTS parmi les catégories suivantes:

  • 12-15 crédits ECTS en Modules technico-scientifiques (TSM)
    Les modules TSM vous transmettent une compétence technique spécifique à votre orientation et complètent les modules de spécialisation décentralisés.
  • 9-12 crédits ECTS en Bases théoriques élargies (FTP)
    Les modules FTP traitent de bases théoriques telles que les mathématiques élevées, la physique, la théorie de l’information, la chimie, etc., vous permettant d’étendre votre profondeur scientifique abstraite et de contribuer à créer le lien important entre l’abstraction et l’application dans le domaine de l’innovation.
  • 6-9 crédits ECTS en Modules contextuels (CM)
    Les modules CM vous transmettent des compétences supplémentaires dans des domaines tels que la gestion des technologies, la gestion d’entreprise, la communication, la gestion de projets, le droit des brevets et des contrats, etc.

Le descriptif de module (download pdf) contient le détail des langues pour chaque module selon les catégories suivantes:

  • leçons
  • documentation
  • examen 
Multimodal Recommendation Systems and Complex Networks (TSM_DataAnaCla)

 

The module will address the theoretical aspects behind the realisation of Recommendation Systems and will allow students to practice over different use case scenarios. In particular it will address the following RecSys approaches: 

  • Traditional and Machine Learning based recommendation
  • Deep Leanrning based recommendation
  • Complex networks based recommendation.


 

 

 

 

Compétences préalables

 

  • Machine Learning and Data Mining, 
  • Python programming
  • basic calculus, linear algebra and statistics concepts

 

Objectifs d'apprentissage

Students understand the theoretical aspects behind the realization of Recommendation Systems and they will learn how to build them over different use case scenarios.

They will learn how recommendation systems work, focusing on three different approaches:

  • Traditional and Machine Learning based recommendation
  • Deep Leanrning based recommendation
  • Complex networks based recommendation.

They will learn how to deal with classical recommendation challenges like imbalanced data set problems, cold-start problems, and long tail problems. Additionally they will learn how to evaluate recommendation systems.

They will learn how to deal with complex networks and how to exploit network extracted information to enhance recommendation solutions.

Finally they will learn how to build multimodal recommendation systems exploiting social networks metrics and dynamics in order to deal with content spread and users engagement.

They know the current research directions within these domains.

They can reuse the material acquired in this course in their own working environment and apply them to solve their specific problems

Catégorie de module

 

The content of the module includes 3 main topics:

Complex Networks:

  • Network Elements (Handling Networks in Code,  Density and Sparsity, Subnetworks, Degree, Multilayer and Temporal Networks,  Network Representations)
  • Network measures (Hubs, Centrality Measures, Centrality Distributions, The Friendship Paradox, Ultra-Small Worlds, Robustness, Core Decomposition, Transitivity, Similarity)
  • Network models (Lattice, Random Networks, Small Worlds, Configuration Model, Preferential Attachment, Other Preferential Models)
  • Community Detection
  • Dynamic models (Ideas, Information, Influence, Epidemic Spreading, Opinion Dynamics,  Search)
  • Social Media as Networks (es. Twitter, Facebook and Reddit)

 

Recommendation Systems:

  • Traditional and Machine Learning based Recommendation Systems (Collaborative Filtering, Content Based, Knowledge Based, Hybrid)
  • Deep Learning based Recommendation Systems
  • Reinforcement Learning based Recommendation Systems
  • Evaluation of Recommendation Systems
  • Handling challenges in Recommendation Systems (imbalanced data set problems, cold-start problems, long tail problems)

Multimodal Systems for Recommendation:

  • Complex Networks/Social Networks integration
  • Use case: Recommendation Systems for Social Networks content spread and users engagement.

 

 

 

Méthodes d'enseignement et d'apprentissage

Problem based learning. During the lesson the lecturer will introduce real world problems and the class will try to solve them together. 

The lecturer will support the problem solving process, introducing new concepts and tools, as required. 

Practical work will complement the theory, so that students can put in practice the studied arguments.

Bibliographie

Lecture slides, references to internet resources and books

Télécharger le descriptif complet

Retour