ranksvm

https://www.cs.cornell.edu/people/tj/publications/joachims_02c.pdf

gbrank

https://www.cc.gatech.edu/~zha/papers/fp086-zheng.pdf

LambdaMART

https://www.microsoft.com/en-us/research/publication/from-ranknet-to-lambdarank-to-lambdamart-an-overview/

特征

  • Web graph - PageRank etc.
  • Document statistics - the number of words in various fields etc.
  • Document classifier - spam, adult, language, main topic, quality, type of page etc.
  • Query Features - terms number, query ctr etc.
  • Text match - title body abstract keyword etc.
  • Topical matching - topic level similarity etc.
  • Click - derived from log etc.
  • Time - the freshness of a page etc.