Personalized Web Search

TL;DR

This paper introduces a novel personalized web search method using a multi-armed bandit approach to model user behavior, improving ranking performance by effectively generalizing session data across users and adapting to new URLs and users.

Contribution

It presents a new multi-armed bandit based algorithm for personalization that captures both long-term and short-term user behaviors, outperforming existing bandit algorithms.

Findings

Improved search ranking performance over default methods.

Effective generalization of session information across users.

Outperforms several popular multi-armed bandit algorithms.

Abstract

Personalization is important for search engines to improve user experience. Most of the existing work do pure feature engineering and extract a lot of session-style features and then train a ranking model. Here we proposed a novel way to model both long term and short term user behavior using Multi-armed bandit algorithm. Our algorithm can generalize session information across users well, and as an Explore-Exploit style algorithm, it can generalize to new urls and new users well. Experiments show that our algorithm can improve performance over the default ranking and outperforms several popular Multi-armed bandit algorithms.