Interactive Event Sifting using Bayesian Graph Neural Networks

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Nascimento, José, Jacobs, Nathan, Rocha, Anderson
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912062509154304
author Nascimento, José
Jacobs, Nathan
Rocha, Anderson
author_facet Nascimento, José
Jacobs, Nathan
Rocha, Anderson
contents Forensic analysts often use social media imagery and texts to understand important events. A primary challenge is the initial sifting of irrelevant posts. This work introduces an interactive process for training an event-centric, learning-based multimodal classification model that automates sanitization. We propose a method based on Bayesian Graph Neural Networks (BGNNs) and evaluate active learning and pseudo-labeling formulations to reduce the number of posts the analyst must manually annotate. Our results indicate that BGNNs are useful for social-media data sifting for forensics investigations of events of interest, the value of active learning and pseudo-labeling varies based on the setting, and incorporating unlabelled data from other events improves performance.
format Preprint
id arxiv_https___arxiv_org_abs_2410_05359
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Interactive Event Sifting using Bayesian Graph Neural Networks
Nascimento, José
Jacobs, Nathan
Rocha, Anderson
Machine Learning
Social and Information Networks
Forensic analysts often use social media imagery and texts to understand important events. A primary challenge is the initial sifting of irrelevant posts. This work introduces an interactive process for training an event-centric, learning-based multimodal classification model that automates sanitization. We propose a method based on Bayesian Graph Neural Networks (BGNNs) and evaluate active learning and pseudo-labeling formulations to reduce the number of posts the analyst must manually annotate. Our results indicate that BGNNs are useful for social-media data sifting for forensics investigations of events of interest, the value of active learning and pseudo-labeling varies based on the setting, and incorporating unlabelled data from other events improves performance.
title Interactive Event Sifting using Bayesian Graph Neural Networks
topic Machine Learning
Social and Information Networks
url https://arxiv.org/abs/2410.05359