A Contextual-Bandit Approach to Online Learning to Rank for Relevance and Diversity.
Chang LiHaoyun FengMaarten de RijkePublished in: CoRR (2019)
Keyphrases
- learning to rank
- information retrieval
- learning to rank algorithms
- balancing exploration and exploitation
- relevance judgments
- query dependent
- contextual bandit
- ranking functions
- loss function
- ranking models
- test collection
- upper confidence bound
- ranking list
- evaluation measures
- ranking svm
- document retrieval
- collaborative filtering
- direct optimization
- supervised learning
- relevance feedback
- information extraction
- normalized discounted cumulative gain
- retrieval systems
- benchmark datasets
- web documents
- news recommendation
- web search
- reinforcement learning