D2RLIR : an improved and diversified ranking function in interactive recommendation systems based on deep reinforcement learning.
Vahid BaghiSeyed Mohammad Seyed MotehayeriAli MoeiniRooholah AbedianPublished in: CoRR (2021)
Keyphrases
- recommendation systems
- ranking functions
- web search
- reinforcement learning
- learning to rank
- supervised learning
- collaborative filtering
- web search engines
- search engine
- document retrieval
- ranking algorithm
- user preferences
- user feedback
- real valued
- recommender systems
- collaborative filtering recommendation algorithm
- ranking scheme
- scoring function
- recommendation algorithms
- recommendation quality
- relevance judgments
- ranking svm
- user profiles
- language model
- feature set
- multi class
- database systems