| Titre : | Systèmes de recommandation basés sur lestechniques del’apprentissage profond |
| Auteurs : | Latreche Abdelkrim, Directeur de thèse ; HAMEL;Wassila, Acteur ; HALIMI Amel Ikram, Autre |
| Type de document : | texte imprimé |
| Editeur : | univ DR taher moulay saida, 2023/2024 |
| Format : | 98 p / ill / 29 cm |
| Accompagnement : | CD rom |
| Langues: | Français |
| Catégories : | |
| Mots-clés: | Apprentissage profond ; Les Systèmes de recommandation ; Implémentation et évaluation |
| Résumé : |
In the current context of overload caused by the large volume of accessible digital data, recommendation systems make it possible to guide the user in their learning, shopping, leisure activities, watching films, reading, etc..., by suggesting personalized items and providing users with suggestions that meet their requirements. To do this, they predict his preferences relative to items that he has not yet evaluated. Among the classic recommendation approaches, collaborative filtering (CF) is the most important and widely used method which relies on data collected through user feedback, usually in the form of a rating matrix, and attempt to discover relevant information to characterize and predict user tastes. In this dissertation, our work is part of the application of deep learning techniques in collaborative recommendation systems. Precisely, the objective of our work is to implement and compare various methods based on deep learning, such as: Co-occurrence CNN for recommendation (CoCNN), CNN for recommendation (CNN), neural collaborative filtering ( NCF) and methods based on machine learning, such as: k nearest neighbors (KNN). We compared these models to determine which method is most appropriate for modeling complex user-element interaction. |
| Note de contenu : |
Chapitre I : Apprentissage profond (Deep learning) Chapitre II: Les Systèmes de recommandation Chapitre III: Implémentation et évaluation |
Exemplaires
| Code-barres | Cote | Support | Localisation | Section | Disponibilité |
|---|---|---|---|---|---|
| aucun exemplaire |
Documents numériques (1)
Systèmes de recommandation basés sur lestechniques del’apprentissage profond URL |
