A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection Paradigm

Multisource and multimodal data fusion plays a pivotal role in large-scale artificial intelligence applications involving big data. However, the choice of fusion strategies for different scenarios is often based on experimental comparisons, which leads to increased computational costs during model t...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziqi Liu, Ziqiao Yin, Zhilong Mi, Binghui Guo, Zhiming Zheng
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/8/1218
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850144742872973312
author Ziqi Liu
Ziqiao Yin
Zhilong Mi
Binghui Guo
Zhiming Zheng
author_facet Ziqi Liu
Ziqiao Yin
Zhilong Mi
Binghui Guo
Zhiming Zheng
author_sort Ziqi Liu
collection DOAJ
description Multisource and multimodal data fusion plays a pivotal role in large-scale artificial intelligence applications involving big data. However, the choice of fusion strategies for different scenarios is often based on experimental comparisons, which leads to increased computational costs during model training and suboptimal performance during testing. In this paper, we present a theoretical analysis of early fusion, late fusion, and gradual fusion methods. We derive equivalence conditions between early and late fusions within the framework of generalized linear models. Moreover, we analyze the failure conditions of early fusion in the presence of nonlinear feature-label relationships. Furthermore, we propose an approximate equation for evaluating the accuracy of early and late fusion methods as a function of sample size, feature quantity, and modality number. We also propose a critical sample size threshold at which the performance dominance of early fusion and late fusion models undergoes a reversal. Finally, we introduce a fusion method selection paradigm for selecting the most appropriate fusion method prior to task execution and demonstrate its effectiveness through extensive numerical experiments. Our theoretical framework is expected to solve the problems of computational and resource costs in model construction, improving the scalability and efficiency of data fusion methods.
format Article
id doaj-art-e8a03daa3b6a44929a37ef873e2eec1a
institution OA Journals
issn 2227-7390
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-e8a03daa3b6a44929a37ef873e2eec1a2025-08-20T02:28:15ZengMDPI AGMathematics2227-73902025-04-01138121810.3390/math13081218A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection ParadigmZiqi Liu0Ziqiao Yin1Zhilong Mi2Binghui Guo3Zhiming Zheng4School of Mathematical Sciences, Beihang University, Beijing 100191, ChinaKey Laboratory of Mathematics, Informatics and Behavioral Semantics and State, Key Laboratory of Software Development Environment, Beihang University, Beijing 100191, ChinaKey Laboratory of Mathematics, Informatics and Behavioral Semantics and State, Key Laboratory of Software Development Environment, Beihang University, Beijing 100191, ChinaKey Laboratory of Mathematics, Informatics and Behavioral Semantics and State, Key Laboratory of Software Development Environment, Beihang University, Beijing 100191, ChinaKey Laboratory of Mathematics, Informatics and Behavioral Semantics and State, Key Laboratory of Software Development Environment, Beihang University, Beijing 100191, ChinaMultisource and multimodal data fusion plays a pivotal role in large-scale artificial intelligence applications involving big data. However, the choice of fusion strategies for different scenarios is often based on experimental comparisons, which leads to increased computational costs during model training and suboptimal performance during testing. In this paper, we present a theoretical analysis of early fusion, late fusion, and gradual fusion methods. We derive equivalence conditions between early and late fusions within the framework of generalized linear models. Moreover, we analyze the failure conditions of early fusion in the presence of nonlinear feature-label relationships. Furthermore, we propose an approximate equation for evaluating the accuracy of early and late fusion methods as a function of sample size, feature quantity, and modality number. We also propose a critical sample size threshold at which the performance dominance of early fusion and late fusion models undergoes a reversal. Finally, we introduce a fusion method selection paradigm for selecting the most appropriate fusion method prior to task execution and demonstrate its effectiveness through extensive numerical experiments. Our theoretical framework is expected to solve the problems of computational and resource costs in model construction, improving the scalability and efficiency of data fusion methods.https://www.mdpi.com/2227-7390/13/8/1218data fusionequivalency analysismodel accuracy evaluationcritical sample size threshold evaluationmethod selection paradigm
spellingShingle Ziqi Liu
Ziqiao Yin
Zhilong Mi
Binghui Guo
Zhiming Zheng
A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection Paradigm
Mathematics
data fusion
equivalency analysis
model accuracy evaluation
critical sample size threshold evaluation
method selection paradigm
title A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection Paradigm
title_full A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection Paradigm
title_fullStr A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection Paradigm
title_full_unstemmed A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection Paradigm
title_short A Comparative Analysis of Three Data Fusion Methods and Construction of the Fusion Method Selection Paradigm
title_sort comparative analysis of three data fusion methods and construction of the fusion method selection paradigm
topic data fusion
equivalency analysis
model accuracy evaluation
critical sample size threshold evaluation
method selection paradigm
url https://www.mdpi.com/2227-7390/13/8/1218
work_keys_str_mv AT ziqiliu acomparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT ziqiaoyin acomparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT zhilongmi acomparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT binghuiguo acomparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT zhimingzheng acomparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT ziqiliu comparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT ziqiaoyin comparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT zhilongmi comparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT binghuiguo comparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm
AT zhimingzheng comparativeanalysisofthreedatafusionmethodsandconstructionofthefusionmethodselectionparadigm