BeBS: Behavior-Biased Search Result Summary Generation

Improving Search Result Summaries By Using Searcher Behavior Data

This page provides source code, data, and the game rules for the paper:

Improving Search Result Summaries By Using Searcher Behavior Data.
Mikhail Ageev, Dmitry Lagun, and Eugene Agichtein.
In Proceedings of the 36th annual international ACM SIGIR conference on Research and development in information retrieval (SIGIR’2013). ACM, Dublin, Ireland.

Full text PDF

To cite out paper:

author = {Mikhail Ageev, Dmitry Lagun, and Eugene Agichtein}
title = {Improving Search Result Summaries By Using Searcher Behavior Data.}
booktitle = {Proceeding of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval({SIGIR} 2013)},
year = {2013},

keywords = {summarization, snippets, user behavior, mouse movements, query log analysis}

Query-biased search result summaries, or “snippets”, help users decide whether a result is relevant for their information need, and have become increasingly important for helping searchers with difficult or ambiguous search tasks. Previously published snippet generation algorithms have been primarily based on selecting document fragments most similar to the query, which does not take into account which parts of the document the searchers actually found useful. We present a new approach to improving result summaries by incorporating post-click searcher behavior data, such as mouse cursor movements and scrolling over the result documents. To achieve this aim, we develop a method for collecting behavioral data with precise association between searcher intent, document examination behavior, and the corresponding document fragments. In turn, this allows us to incorporate page examination behavior signals into a novel Behavior-Biased Snippet generation system (BeBS). By mining searcher examination data, BeBS infers document fragments of most interest to users, and combines this evidence with text-based features to select the most promising fragments for inclusion in the result summary. Our extensive experiments and analysis demonstrate that our method improves the quality of result summaries compared to existing state-of-the-art methods. We believe that this work opens a new direction for improving search result presentation, and we make available the code and the search behavior data used in this study to encourage further research in this area.


  • The source code and supplemental data for user behavior tracking and snippet generation:, 38Mb
  • Full dataset: fine-grained user behavior log, landing pages, and snippets (PostgreSQL backup): snip.bkp.bz2, 2.5Gb
  • The list of questions used in the game: questions.txt

This work was supported by the National Science Foundation Award Number IIS 1018321
“Modeling and Inferring Searcher Intent by Mining User Interactions”