Fast Pairwise Query Selection for Large-Scale Active Learning to Rank.
Buyue QianXiang WangJun WangHongfei LiNan CaoWeifeng ZhiIan DavidsonPublished in: ICDM (2013)
Keyphrases
- learning to rank
- pairwise
- ranking svm
- loss function
- query dependent
- relevance judgments
- ranking functions
- retrieval systems
- ranking list
- query suggestion
- evaluation measures
- information retrieval
- ranking models
- learn a ranking function
- document retrieval
- test collection
- direct optimization
- evaluation metrics
- data sources
- ranking algorithm
- multi class
- directly optimize
- user queries
- normalized discounted cumulative gain
- query processing
- learning to rank algorithms
- similarity measure
- training data
- support vector
- average precision
- web search
- clickthrough data
- search engine
- collaborative filtering
- query expansion
- data sets