Unmasking social bots: how confident are we?

Abstract Social bots remain a major vector for spreading disinformation on social media and a menace to the public. Despite the progress made in developing multiple sophisticated social bot detection algorithms and tools, bot detection remains a challenging, unsolved problem that is fraught with unc...

Full description

Saved in:
Bibliographic Details
Main Authors: James Giroux, Gangani Ariyarathne, Alexander C. Nwala, Cristiano Fanelli
Format: Article
Language:English
Published: SpringerOpen 2025-03-01
Series:EPJ Data Science
Subjects:
Online Access:https://doi.org/10.1140/epjds/s13688-025-00536-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849761604827086848
author James Giroux
Gangani Ariyarathne
Alexander C. Nwala
Cristiano Fanelli
author_facet James Giroux
Gangani Ariyarathne
Alexander C. Nwala
Cristiano Fanelli
author_sort James Giroux
collection DOAJ
description Abstract Social bots remain a major vector for spreading disinformation on social media and a menace to the public. Despite the progress made in developing multiple sophisticated social bot detection algorithms and tools, bot detection remains a challenging, unsolved problem that is fraught with uncertainty due to the heterogeneity of bot behaviors, training data, and detection algorithms. Detection models often disagree on whether to label the same account as bot or human-controlled. However, they do not provide any measure of uncertainty to indicate how much we should trust their results. We propose to address both bot detection and the quantification of uncertainty at the account level — a novel feature of this research. This dual focus is crucial as it allows us to leverage additional information related to the quantified uncertainty of each prediction, thereby enhancing decision-making and improving the reliability of bot classifications. Specifically, our approach facilitates targeted interventions for bots when predictions are made with high confidence and suggests caution (e.g., gathering more data) when predictions are uncertain.
format Article
id doaj-art-4ee4e37ff51b4d6a8de989c04b59e4b2
institution DOAJ
issn 2193-1127
language English
publishDate 2025-03-01
publisher SpringerOpen
record_format Article
series EPJ Data Science
spelling doaj-art-4ee4e37ff51b4d6a8de989c04b59e4b22025-08-20T03:05:57ZengSpringerOpenEPJ Data Science2193-11272025-03-0114111910.1140/epjds/s13688-025-00536-yUnmasking social bots: how confident are we?James Giroux0Gangani Ariyarathne1Alexander C. Nwala2Cristiano Fanelli3Department of Data Science, William & MaryDepartment of Data Science, William & MaryDepartment of Data Science, William & MaryDepartment of Data Science, William & MaryAbstract Social bots remain a major vector for spreading disinformation on social media and a menace to the public. Despite the progress made in developing multiple sophisticated social bot detection algorithms and tools, bot detection remains a challenging, unsolved problem that is fraught with uncertainty due to the heterogeneity of bot behaviors, training data, and detection algorithms. Detection models often disagree on whether to label the same account as bot or human-controlled. However, they do not provide any measure of uncertainty to indicate how much we should trust their results. We propose to address both bot detection and the quantification of uncertainty at the account level — a novel feature of this research. This dual focus is crucial as it allows us to leverage additional information related to the quantified uncertainty of each prediction, thereby enhancing decision-making and improving the reliability of bot classifications. Specifically, our approach facilitates targeted interventions for bots when predictions are made with high confidence and suggests caution (e.g., gathering more data) when predictions are uncertain.https://doi.org/10.1140/epjds/s13688-025-00536-yUncertainty quantificationBayesian neural networkSocial mediaBot detection
spellingShingle James Giroux
Gangani Ariyarathne
Alexander C. Nwala
Cristiano Fanelli
Unmasking social bots: how confident are we?
EPJ Data Science
Uncertainty quantification
Bayesian neural network
Social media
Bot detection
title Unmasking social bots: how confident are we?
title_full Unmasking social bots: how confident are we?
title_fullStr Unmasking social bots: how confident are we?
title_full_unstemmed Unmasking social bots: how confident are we?
title_short Unmasking social bots: how confident are we?
title_sort unmasking social bots how confident are we
topic Uncertainty quantification
Bayesian neural network
Social media
Bot detection
url https://doi.org/10.1140/epjds/s13688-025-00536-y
work_keys_str_mv AT jamesgiroux unmaskingsocialbotshowconfidentarewe
AT ganganiariyarathne unmaskingsocialbotshowconfidentarewe
AT alexandercnwala unmaskingsocialbotshowconfidentarewe
AT cristianofanelli unmaskingsocialbotshowconfidentarewe