Learning Dense Models of Query Similarity from User Click Logs

  title={Learning Dense Models of Query Similarity from User Click Logs},
  author={Fabio De Bona and Stefan Riezler and Keith B. Hall and Massimiliano Ciaramita and Amac Herdagdelen and Maria Holmqvist},
The goal of this work is to integrate query similarity metrics as features into a dense model that can be trained on large amounts of query log data, in order to rank query rewrites. We propose features that incorporate various notions of syntactic and semantic similarity in a generalized edit distance framework. We use the implicit feedback of user clicks on search results as weak labels in training linear ranking models on large data sets. We optimize different ranking objectives in a… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.

Similar Papers

Loading similar papers…