Context Models For Web Search Personalization
This work addresses web search personalization for users, but it is incremental as it applies existing methods to a specific challenge.
The authors tackled the Yandex Personalized Web Search Challenge by using historical search logs and over 100 features to train neural net and tree-based models for personalizing top-N document rankings, achieving an NDCG@10 of 0.80476 and placing 4th out of 194 teams.
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.