Context Models For Web Search Personalization

Maksims Volkovs

Context Models For Web Search Personalization

Information Retrieval 2015-02-03 v1

Authors: Maksims Volkovs

Abstract

We present our solution to the Yandex Personalized Web Search Challenge. The aim of this challenge was to use the historical search logs to personalize top-N document rankings for a set of test users. We used over 100 features extracted from user- and query-depended contexts to train neural net and tree-based learning-to-rank and regression models. Our final submission, which was a blend of several different models, achieved an NDCG@10 of 0.80476 and placed 4'th amongst the 194 teams winning 3'rd prize.

Keywords

information retrieval

Cite

@article{arxiv.1502.00527,
  title  = {Context Models For Web Search Personalization},
  author = {Maksims Volkovs},
  journal= {arXiv preprint arXiv:1502.00527},
  year   = {2015}
}

Related papers

View all related →