Self-supervised Information Retrieval Trained from Self-generated Sets of Queries and Relevant Documents.
Gianluca MoroLorenzo ValgimigliAlex RossiCristiano CasadeiAndrea MontefioriPublished in: SISAP (2022)
Keyphrases
- bayesian networks
- relevant documents
- user queries
- information retrieval
- query terms
- test collection
- search queries
- information retrieval systems
- original query
- result set
- relevance judgments
- retrieval effectiveness
- document collections
- document retrieval
- query expansion
- query specific
- search engine
- web search engines
- retrieval systems
- retrieved documents
- probabilistic retrieval models
- documents retrieved
- query processing
- keywords
- ad hoc retrieval
- ranked list
- collection selection
- retrieve documents
- retrieval model
- relevance feedback
- number of relevant documents
- recall oriented
- pseudo relevance feedback
- query logs
- document rankings
- structured queries
- language model
- query suggestion
- learning to rank
- short queries
- result list
- document ranking
- relevance ranking
- vector space model
- relevance assessments
- term selection
- information extraction
- training set
- relevance model
- query refinement
- anchor text
- language modeling
- current search engines
- term frequency
- focused retrieval
- web search
- distributed information retrieval
- meta search
- document set