FedBID and FedDocs: A Dataset and System for Federated Document Analysis

Daniel Perazzo, Thiago de Souza, Pietro Masur, Eduardo de Amorim, Pedro de Oliveira, Kelvin Cunha, Lucas Maggi, Francisco Simões, Francisco Simões, Veronica Teichrieb, Lucas Kirsten

2023

Abstract

Data privacy has recently become one of the main concerns for society and machine learning researchers. The question of privacy led to research in privacy-aware machine learning and, amongst many other techniques, one solution gaining ground is federated learning. In this machine learning paradigm, data does not leave the user’s device, with training happening on it and aggregated in a remote server. In this work, we present, to our knowledge, the first federated dataset for document classification: FedBID. To demonstrate how this dataset can be used for evaluating different techniques, we also developed a system, FedDocs, for federated learning for document classification. We demonstrate the characteristics of our federated dataset, along with different types of distributions possible to be created with our dataset. Finally, we analyze our system, FedDocs, in our dataset, FedBID, in multiple different scenarios. We analyze a federated setting with balanced categories, a federated setting with unbalanced classes, and, finally, simulating a siloed federated training. We demonstrate that FedBID can be used to analyze a federated learning algorithm. Finally, we hope the FedBID dataset allows more research in federated document classification. The dataset is available in https://github.com/voxarlabs/FedBID.

Paper Citation

in Harvard Style

Perazzo D., de Souza T., Masur P., de Amorim E., de Oliveira P., Cunha K., Maggi L., Simões F., Teichrieb V. and Kirsten L. (2023). FedBID and FedDocs: A Dataset and System for Federated Document Analysis. In Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP; ISBN 978-989-758-634-7, SciTePress, pages 551-558. DOI: 10.5220/0011658700003417

in Bibtex Style

@conference{visapp23,
author={Daniel Perazzo and Thiago de Souza and Pietro Masur and Eduardo de Amorim and Pedro de Oliveira and Kelvin Cunha and Lucas Maggi and Francisco Simões and Veronica Teichrieb and Lucas Kirsten},
title={FedBID and FedDocs: A Dataset and System for Federated Document Analysis},
booktitle={Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP},
year={2023},
pages={551-558},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011658700003417},
isbn={978-989-758-634-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP
TI - FedBID and FedDocs: A Dataset and System for Federated Document Analysis
SN - 978-989-758-634-7
AU - Perazzo D.
AU - de Souza T.
AU - Masur P.
AU - de Amorim E.
AU - de Oliveira P.
AU - Cunha K.
AU - Maggi L.
AU - Simões F.
AU - Teichrieb V.
AU - Kirsten L.
PY - 2023
SP - 551
EP - 558
DO - 10.5220/0011658700003417
PB - SciTePress