Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced Data

Federated learning is increasingly being considered for sensor-driven human activity recognition, offering advantages in terms of privacy and scalability compared to centralized methods. However, challenges such as feature selection and client imbalanced data persist. In this study, FLP-DS2MOTE-USA...

Full description

Saved in:
Bibliographic Details
Main Authors: Zahraa Khduair Taha, Johnny Koh Siaw Paw, Yaw Chong Tak, Tiong Sieh Kiong, Kumaran Kadirgama, Foo Benedict, Tan Jian Ding, Kharudin Ali, Azher M. Abed
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10614580/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850114263903895552
author Zahraa Khduair Taha
Johnny Koh Siaw Paw
Yaw Chong Tak
Tiong Sieh Kiong
Kumaran Kadirgama
Foo Benedict
Tan Jian Ding
Kharudin Ali
Azher M. Abed
author_facet Zahraa Khduair Taha
Johnny Koh Siaw Paw
Yaw Chong Tak
Tiong Sieh Kiong
Kumaran Kadirgama
Foo Benedict
Tan Jian Ding
Kharudin Ali
Azher M. Abed
author_sort Zahraa Khduair Taha
collection DOAJ
description Federated learning is increasingly being considered for sensor-driven human activity recognition, offering advantages in terms of privacy and scalability compared to centralized methods. However, challenges such as feature selection and client imbalanced data persist. In this study, FLP-DS2MOTE-USA is suggested, a system that integrates federated local preprocessing, adaptive thresholding based on uncertainty symmetry, and a density- sensitive synthetic minority over-sampling approach. Each client preprocesses data locally and employs DS2MOTE for class balancing. On the server side, adaptive thresholding based on uncertainty symmetry is utilized to identify the optimal client for training the global mode. Evaluation on two distinct datasets—Human Activity Recognition with Smartphones and Human Activity Recognition (OpenPose) —reveals that our model outperforms FedAvg, FedSgd, FedSmote, and FedNova, achieving accuracies of 90.57% and 96.58%, respectively. In addition, FLP-DS2MOTE-USA minimizes update size and network overhead on the Human Activity Recognition with Smartphones, while achieving improvements on the OpenPose dataset. Overall, the proposed method not only addresses issues of imbalanced data but also reduces computational complexity via streamlined local preprocessing, and server-side mechanisms ensure client privacy. It outperforms traditional federated learning techniques in both accuracy and efficiency.
format Article
id doaj-art-9e91666ff9044a7a9335d126f99e0044
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-9e91666ff9044a7a9335d126f99e00442025-08-20T02:36:57ZengIEEEIEEE Access2169-35362024-01-011218627718629510.1109/ACCESS.2024.343591010614580Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced DataZahraa Khduair Taha0https://orcid.org/0000-0003-4889-9679Johnny Koh Siaw Paw1Yaw Chong Tak2https://orcid.org/0000-0002-5044-6110Tiong Sieh Kiong3https://orcid.org/0000-0002-4447-262XKumaran Kadirgama4Foo Benedict5Tan Jian Ding6Kharudin Ali7Azher M. Abed8College of Graduate Studies (COGS), Universiti Tenaga Nasional (The Energy University), Kajang, MalaysiaInstitute of Sustainable Energy, Universiti Tenaga Nasional (The Energy University), Kajang, MalaysiaInstitute of Sustainable Energy, Universiti Tenaga Nasional (The Energy University), Kajang, MalaysiaInstitute of Sustainable Energy, Universiti Tenaga Nasional (The Energy University), Kajang, MalaysiaFaculty of Mechanical and Automotive Engineering Technology, Universiti Malaysia Pahang, Pekan, Pahang, MalaysiaEnhance Track Sdn. Bhd., Puchong, Selangor, MalaysiaSchool of Electrical Engineering and Artificial Intelligence, Xiamen University Malaysia, Sepang, Selangor, MalaysiaFaculty of Engineering Technology (Electrical and Automation), University College TATI, Kemaman, Terengganu, MalaysiaCollege of Engineering and Technologies, Al-Mustaqbal University, Babylon, IraqFederated learning is increasingly being considered for sensor-driven human activity recognition, offering advantages in terms of privacy and scalability compared to centralized methods. However, challenges such as feature selection and client imbalanced data persist. In this study, FLP-DS2MOTE-USA is suggested, a system that integrates federated local preprocessing, adaptive thresholding based on uncertainty symmetry, and a density- sensitive synthetic minority over-sampling approach. Each client preprocesses data locally and employs DS2MOTE for class balancing. On the server side, adaptive thresholding based on uncertainty symmetry is utilized to identify the optimal client for training the global mode. Evaluation on two distinct datasets—Human Activity Recognition with Smartphones and Human Activity Recognition (OpenPose) —reveals that our model outperforms FedAvg, FedSgd, FedSmote, and FedNova, achieving accuracies of 90.57% and 96.58%, respectively. In addition, FLP-DS2MOTE-USA minimizes update size and network overhead on the Human Activity Recognition with Smartphones, while achieving improvements on the OpenPose dataset. Overall, the proposed method not only addresses issues of imbalanced data but also reduces computational complexity via streamlined local preprocessing, and server-side mechanisms ensure client privacy. It outperforms traditional federated learning techniques in both accuracy and efficiency.https://ieeexplore.ieee.org/document/10614580/Federated learninglocal preprocessingimbalance datauncertainty symmetry
spellingShingle Zahraa Khduair Taha
Johnny Koh Siaw Paw
Yaw Chong Tak
Tiong Sieh Kiong
Kumaran Kadirgama
Foo Benedict
Tan Jian Ding
Kharudin Ali
Azher M. Abed
Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced Data
IEEE Access
Federated learning
local preprocessing
imbalance data
uncertainty symmetry
title Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced Data
title_full Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced Data
title_fullStr Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced Data
title_full_unstemmed Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced Data
title_short Advances in Federated Learning: Combining Local Preprocessing With Adaptive Uncertainty Symmetry to Reduce Irrelevant Features and Address Imbalanced Data
title_sort advances in federated learning combining local preprocessing with adaptive uncertainty symmetry to reduce irrelevant features and address imbalanced data
topic Federated learning
local preprocessing
imbalance data
uncertainty symmetry
url https://ieeexplore.ieee.org/document/10614580/
work_keys_str_mv AT zahraakhduairtaha advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT johnnykohsiawpaw advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT yawchongtak advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT tiongsiehkiong advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT kumarankadirgama advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT foobenedict advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT tanjianding advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT kharudinali advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata
AT azhermabed advancesinfederatedlearningcombininglocalpreprocessingwithadaptiveuncertaintysymmetrytoreduceirrelevantfeaturesandaddressimbalanceddata