Investigating Bias in Political Search Query Suggestions by Relative Comparison with LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Haak, Fabian, Engelmann, Björn, Kreutz, Christin Katharina, Schaer, Philipp
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916463207514112
author Haak, Fabian
Engelmann, Björn
Kreutz, Christin Katharina
Schaer, Philipp
author_facet Haak, Fabian
Engelmann, Björn
Kreutz, Christin Katharina
Schaer, Philipp
contents Search query suggestions affect users' interactions with search engines, which then influences the information they encounter. Thus, bias in search query suggestions can lead to exposure to biased search results and can impact opinion formation. This is especially critical in the political domain. Detecting and quantifying bias in web search engines is difficult due to its topic dependency, complexity, and subjectivity. The lack of context and phrasality of query suggestions emphasizes this problem. In a multi-step approach, we combine the benefits of large language models, pairwise comparison, and Elo-based scoring to identify and quantify bias in English search query suggestions. We apply our approach to the U.S. political news domain and compare bias in Google and Bing.
format Preprint
id arxiv_https___arxiv_org_abs_2410_23879
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Investigating Bias in Political Search Query Suggestions by Relative Comparison with LLMs
Haak, Fabian
Engelmann, Björn
Kreutz, Christin Katharina
Schaer, Philipp
Information Retrieval
94-02
H.3.3
Search query suggestions affect users' interactions with search engines, which then influences the information they encounter. Thus, bias in search query suggestions can lead to exposure to biased search results and can impact opinion formation. This is especially critical in the political domain. Detecting and quantifying bias in web search engines is difficult due to its topic dependency, complexity, and subjectivity. The lack of context and phrasality of query suggestions emphasizes this problem. In a multi-step approach, we combine the benefits of large language models, pairwise comparison, and Elo-based scoring to identify and quantify bias in English search query suggestions. We apply our approach to the U.S. political news domain and compare bias in Google and Bing.
title Investigating Bias in Political Search Query Suggestions by Relative Comparison with LLMs
topic Information Retrieval
94-02
H.3.3
url https://arxiv.org/abs/2410.23879