Abstract
Facial expression recognition is crucial in analyzing an individual’s emotional state. Ensemble pruning becomes essential to enhance the effectiveness of this recognition by selecting appropriate classifiers from a pool of base classifiers. However, outliers and noise within the classifiers can adversely affect the final recognition results and the generalization performance of the selected subset of classifiers. Additionally, effectively combining accuracy and low redundancy of base classifiers remains a challenging problem that requires further investigation. In this paper, we propose a novel algorithm called Pairwise Dependency-based Robust Ensemble Pruning (PDREP) to address these issues. The PDREP algorithm treats the predicted results of classifiers for sample instances as features of the classifier and evaluates their dependencies between pairs of classifiers using mutual information. By incorporating this dependency measure into the regression-based objective equation, we can assess the redundancy of a subset of base classifiers and prune redundant classifiers. We use the \(\varvec{l}_{\textbf{2},\textbf{1}}\)-norm in PDREP’s objective equation to perform robust classifier pruning while considering the base classifiers’ dependencies and accuracy. Furthermore, we introduce weight control parameters to balance accuracy and dependencies, facilitating the elimination of ineffective and redundant classifiers. The results show significant improvements when evaluating the proposed method’s recognition accuracy on five public face sentiment datasets (FER2013, JAFFE, CK+, RaFD, and KDEF). Specifically, the proposed method achieves 3.15%, 9.39%, 1.72%, 3.70%, and 4.70% higher accuracy than integrating all base classifiers, respectively. Extensive experiments conducted on five facial expression datasets demonstrate that the proposed method consistently outperforms existing state-of-the-art ensemble pruning algorithms in most cases.
Similar content being viewed by others
Data Availability
The datasets employed and analyzed throughout this study are openly accessible through the following repositories:
FER2013 dataset: https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge/data
JAFFE dataset: https://zenodo.org/record/3451524
CK+ dataset: http://vasc.ri.cmu.edu/idb/html/face/facial_expression/
RaFD dataset: http://www.socsci.ru.nl:8180/RaFD2/RaFD
KDEF dataset: https://www.kdef.se/home/aboutKDEF
Change history
08 February 2024
A Correction to this paper has been published: https://doi.org/10.1007/s11042-024-18565-6
References
Lyons MJ, Kamachi M, Gyoba J (2020) Coding Facial Expressions with Gabor Wavelets (IVC Special Issue). https://doi.org/10.5281/zenodo.4029680
Li X et al (2023) Facial expression recognition network with slow convolution and zero-parameter attention mechanism. Optik 283:170892. https://doi.org/10.1016/j.ijleo.2023.170892
Lyons MJ (2021) “Excavating AI” Revisited: Discrediting Misleading Accounts of the JAFFE Dataset. zenodo. https://doi.org/10.5281/zenodo.5147170
Li D, Wen G, Li X, Cai X (2019) Graph-based dynamic ensemble pruning for facial expression recognition. Appl Intell 49:3188–3206. https://doi.org/10.1007/s10489-019-01435-2
Li D, Wen G (2018) Mrmr-based ensemble pruning for facial expression recognition. Multimed Tools Appl. https://doi.org/10.1007/s11042-017-5105-z
Li D et al (2018) Rtcrelief-f: an effective clustering and ordering-based ensemble pruning algorithm for facial expression recognition. Knowl Inf Syst 59. https://doi.org/10.1007/s10115-018-1176-z
Zhenlei W et al (2021) Ensemble selection with joint spectral clustering and structural sparsity. Pattern Recognit 119:108061. https://doi.org/10.1016/j.patcog.2021.108061
Huang S et al (2023) Cslsep: an ensemble pruning algorithm based on clustering soft label and sorting for facial expression recognition. Multimed Syst 29:1463–1479. https://doi.org/10.1007/s00530-023-01062-5
Dai Q, Ye R, Liu Z (2017) Considering diversity and accuracy simultaneously for ensemble pruning. Appl Soft Comput 58:75–91. https://doi.org/10.1016/j.asoc.2017.04.058
Fletcher S, Verma B, Zhang M (2020) A non-specialized ensemble classifier using multi-objective optimization. Neurocomputing 409:93–102. https://doi.org/10.1016/j.neucom.2020.05.029
Lim H, Kim D-W (2021) Pairwise dependence-based unsupervised feature selection. Pattern Recognit 111:107663. https://doi.org/10.1016/j.patcog.2020.107663
Du S, Ma Y, Li S, Ma Y (2017) Robust unsupervised feature selection via matrix factorization. Neurocomputing 241:115–127. https://doi.org/10.1016/j.neucom.2017.02.034
Xia X, Jiang D (2023) Hit-mst: Dynamic facial expression recognition with hierarchical transformers and multi-scale spatiotemporal aggregation. Inform Sci 644:119301. https://doi.org/10.1016/j.ins.2023.119301
Wu X et al (2023) Fer-chc: Facial expression recognition with cross-hierarchy contrast. Appl Soft Comp 110530. https://doi.org/10.1016/j.asoc.2023.110530
Nan Y, Ju J, Hua Q, Zhang H, Wang B (2022) A-mobilenet: An approach of facial expression recognition. Alex Eng J 61:4435–4444. https://doi.org/10.1016/j.aej.2021.09.066
Guo Y et al (2022) Facial expressions recognition with multi-region divided attention networks for smart education cloud applications. Neurocomputing 493:119–128. https://doi.org/10.1016/j.neucom.2022.04.052
Ganaie MA, Hu M, Malik A, Tanveer M, Suganthan P (2022) Ensemble deep learning: A review. Eng Appl Artif Intell 115:105151. https://doi.org/10.1016/j.engappai.2022.105151
Chirra VRR, Uyyala SR, Kolli VKK (2021) Virtual facial expression recognition using deep cnn with ensemble learning. J Ambient Intell Humaniz Comput 1–19. https://doi.org/10.1109/10.1007/s12652-020-02866-3
Rajagopal SD, Ramachandran B (2022) 3d face expression recognition with ensemble deep learning exploring congruent features among expressions. Comput Intell 38:345–365. https://doi.org/10.1111/coin.12498
Tsai K-Y, Tsai Y-W, Lee Y-C, Ding J-J, Chang RY (2021) Frontalization and adaptive exponential ensemble rule for deep-learning-based facial expression recognition system. Signal Process Image Commun 96:116321. https://doi.org/10.1016/j.image.2021.116321
Jia X, Zhang Y, Powers D, Ali HB (2014) Multi-classifier fusion based facial expression recognition approach. KSII Trans Internet Inform Syst 8. https://doi.org/10.1007/s11042-020-08746-4
Li D, Zhang Z, Wen G (2023) Classifier subset selection based on classifier representation and clustering ensemble. Appl Intell 1–23. https://doi.org/10.1007/s10489-023-04572-x
Roshan E (2021) Development of ensemble learning classification with density peak decomposition-based evolutionary multi-objective optimization. Int J Mach Learn Cybernet 12:1737–1751. https://doi.org/10.1007/s13042-020-01271-8
Ji J, Li J (2023) Tri-objective optimization-based cascade ensemble pruning for deep forest. Pattern Recognit 143:109744. https://doi.org/10.1016/j.patcog.2023.109744
Yang AY, Lv BH, Chen CN, Wu DY, Zheng EZ (2020) Ftbme: feature transferring based multi-model ensemble. Multimed Tools Appl 79:18767–18799. https://doi.org/10.1007/s11042-020-08746-4
Shen X, Dai Q, Zhu G, Zhang J (2023) Dynamic ensemble pruning algorithms fusing meta-learning with heuristic parameter optimization for time series prediction. Exp Syst Appl 225:120148. https://doi.org/10.1016/j.eswa.2023.120148
Martínez-Muñoz G, Suárez A (2006) Pruning in ordered bagging ensembles https://doi.org/10.1145/1143844.1143921
Guo H et al (2018) Margin & diversity based ordering ensemble pruning. Neurocomputing 275:237–246. https://doi.org/10.1016/j.neucom.2017.06.052
Zhang H, Wu S, Zhang X, Han L, Zhang Z (2022) Slope stability prediction method based on the margin distance minimization selective ensemble. CATENA 212:106055. https://doi.org/10.1016/j.catena.2022.106055
Lv S, Peng L, Hu H, Wang L (2022) Effective machine learning model combination based on selective ensemble strategy for time series forecasting. Inform Sci 612:994–1023. https://doi.org/10.1016/j.ins.2022.09.002
Bakkouri I, Afdel K (2023) Mlca2f: Multi-level context attentional feature fusion for covid-19 lesion segmentation from ct scans. Sig Image Video Process 17:1181–1188. https://doi.org/10.3837/tiis.2014.01.012
Bakkouri I, Karim A (2020) Computer-aided diagnosis (cad) system based on multi-layer feature fusion network for skin lesion recognition in dermoscopy images. Multimed Tools Appl 79:20483–20518. https://doi.org/10.1007/s11042-019-07988-1
Afonso MV, Bioucas-Dias JM, Figueiredo MAT (2011) An augmented lagrangian approach to the constrained optimization formulation of imaging inverse problems. IEEE Trans Image Process 20:681–695. https://doi.org/10.1109/TIP.2010.2076294
Goodfellow IJ, Erhan D, Carrier PL et al. (2015) Challenges in representation learning: a report on three machine learning contests. Neural Networks 64:59–63. https://doi.org/10.1016/j.neunet.2014.09.005
Langner O, Dotsch R, Bijlstra G, Wigboldus DHJ, Hawk ST, van Knippenberg A (2010) Presentation and validation of the Radboud Faces Database. Cognition and Emotion 24(8)1377–1388. https://doi.org/10.1080/02699930903485076
Goeleven E, Raedt RD, Leyman L, Verschuere B (2008) The karolinska directed emotional faces: A validation study. Cogn Emot 22:1094–1118. https://doi.org/10.1080/02699930701626582
Dai Q (2013) A novel ensemble pruning algorithm based on randomized greedy selective strategy and ballot. Neurocomputing 122:258–265. https://doi.org/10.1016/j.neucom.2013.06.026
Yin X-C, Huang K, Yang C, Hao H-W (2014) Convex ensemble learning with sparsity and diversity. Inf Fusion 20:49–59. https://doi.org/10.1016/j.inffus.2013.11.003
Kuncheva LI, Whitaker CJ (2003) Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy. Mach Learn 51:181–207. https://doi.org/10.1023/A:1022859003006
Hou C, Xia Y, Xu Z, Sun J (2016). Learning classifier competence based on graph for dynamic classifier selection. https://doi.org/10.1109/FSKD.2016.7603343
Woods K, Kegelmeyer W, Bowyer K (1997) Combination of multiple classifiers using local accuracy estimates. IEEE Trans Pattern Anal Mach Intell 19:405–410. https://doi.org/10.1109/34.588027
Didaci L, Giacinto G, Roli F, Marcialis GL (2005) A study on the performances of dynamic classifier selection based on local accuracy estimation. Pattern Recogn 38:2188–2191. https://doi.org/10.1016/j.patcog.2005.02.010
Markatopoulou F, Tsoumakas G, Vlahavas I (2014) Dynamic ensemble pruning based on multi-label classification. Neurocomputing 150:501–512. https://doi.org/10.1016/j.neucom.2014.07.063
Yang Z, Lu H, Yu Q (2022) Critical independent sets of könig-egerváry graphs. Discret Appl Math 318:1–5. https://doi.org/10.1016/j.dam.2022.04.014
Acknowledgements
This work was supported by the Science and Technology Plan Project of Guizhou Province (Qiankehe Platform Talents [2018] 5781) and was supported by the Guizhou Provincial Science and Technology Plan Project (Qiankehe Support [2023] General 251).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors have no competing interests to declare that are relevant to the content of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The original online version of this article was revised: The original publication of this article contains incorrect details of references [1], [3], [34], and [35] and incorrect Data availability statement.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Chen, X., Li, D., Tang, Y. et al. Pairwise dependency-based robust ensemble pruning for facial expression recognition. Multimed Tools Appl 83, 37089–37117 (2024). https://doi.org/10.1007/s11042-023-16756-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-16756-1