Loading...

Recommender Systems in Education

Arab Jabal Amel, Setareh | 2023

51 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56669 (02)
  4. University: Sharif University of Technology
  5. Department: Mathematical Sciences
  6. Advisor(s): Moghadasi, Reza
  7. Abstract:
  8. With the increasing advancement of technology and available information, finding suitable content for users is one of the significant challenges. Recommender systems have emerged as an effective solution to confront these challenges. Reinforcement learning can be employed as an effective approach to designing and implementing recommender systems. However, there are various challenges and complexities in the context of reinforcement learning applied to recommendation systems. In this context, Online users are treated as the environment, and concepts like reward functions and environment dynamics are not clearly defined, complicating the reinforcement learning process. In this thesis, a model-based reinforcement learning framework for recommendation systems is investigated. In this framework, a Generative Adversarial Network is utilized to simulate user behavior and learn their reward function. Using this simulated user model, a novel algorithm called Cascading DQN is developed to provide a combined recommendation policy. The main goal of the recommendation system is to provide users with a set of items. The use of this innovative algorithm leads to a significant reduction in computational complexity, making the system more efficient and faster in operation, guiding users better to their preferred items. The experimental results with real-world data demonstrate that the introduced Generative Adversarial User Model outperforms traditional approaches in explaining user behavior, and the reinforcement learning policy based on this model leads to higher long-term rewards for users and an overall improvement in the click-through rate of the recommender system
  9. Keywords:
  10. Recommender System ; Reinforcement Learning ; Deep Learning ; Simulators ; Deep Neural Networks ; Generative Neural Networks ; Cascading Deep Q_Networks ; Education Systems

 Digital Object List

 Bookmark

No TOC