Open Access
Research Article
Optimized stacked deep ensemble classifier using hyperparameter tuning for facial expression recognition
K. V. Krishna Kishorekishorekvk1@yahoo.comDepartment of Computer Science Engineering Vignan’s Foundation for Science, Technology & Research (Deemed to be University)Guntur, Andhra Pradesh, 522213, IndiaView full profile → , *Venkata Rami Reddy ChirraCorresponding authorchvrr58@gmail.comSchool of Computer Science Engineering VIT-AP UniversityAmaravathi, Andhra Pradesh, 522241, IndiaView full profile → , U. Srinivasulu Reddyusreddy@nitt.eduDepartment of Computer Applications National Institute of TechnologyTiruchirappalli, Tamil Nadu, 620015, IndiaView full profile →
* Corresponding author · click or hover a name for details
- Received:
- 10 Dec 2024
- Published Online:
- 31 Aug 2025
- Article type:
- Research Article
- Language:
- EN
- Article no.:
- JIOS-2029
- Pages:
- 2005–2015
Abstract
Despite significant advancements in FER methodologies, real-time implementation remains challenging due to intra-class variations such as pose, illumination, age, and skin color. These variations cause Single Convolutional Neural Network (CNN) architectures to suffer from high variance, negatively impacting system performance. To address these challenges, this study introduces a method to mitigate intra-class variations by integrating multiple CNN architectures and evaluating them on the FER-2013 dataset. The objective is to develop a Stacked Deep Ensemble Classifier (SDEC) that selects the best-performing models for ensemble learning, reducing variance and enhancing accuracy. The approach leverages both stacking and voting ensemble techniques to improve prediction reliability. Four high-performing base models were designed and developed for this work. In the stacking ensemble, the outputs of these base models were combined and processed through a meta-classifier to generate the final prediction. In the voting ensemble, each model independently predicted class probabilities, which were then aggregated, with the final class determined using the argmax function. The proposed models were evaluated on the FER-2013, achieving a test accuracy of 70.5% for the stacking ensemble and 70.4% for the voting ensemble.
Keywords
Subject Classifications
68T0768T42
References
[1] T. K. Mohammed, D. N. Vasundhara, S. H. Mehanoor, E. Sreedevi, P. R. Kumar, C. Manihass, and S. F. Baba, “A novel fusion approach for advancement in crime prediction and forecasting using hybridization of ARIMA and recurrent neural networks,” Journal of Information Systems Engineering and Management, vol. 10, no. 38, pp. 404–420 (2025).[2] R. Q. Abdulkadhim, H. S. Abdullah, and M. J. Hadi, “Improving cryptocurrency price prediction through advanced LSTM-based deep learning techniques,” Journal of Statistics and Management Systems, vol. 27, no. 8, pp. 1701–1712 (2024).[3] B. Shilpa, P. R. Kumar, and R. K. Jha, “LoRa DL: A deep learning model for enhancing the data transmission over LoRa using autoencoder,” J. Supercomput., vol. 79, no. 15, pp. 17079–17097 (2023).[4] N. Arivazhagan, K. Somasundaram, G. B. Mohammad, P. R. Kumar, et al., “Cloud-Internet of Health Things (IoHT) task scheduling using hybrid moth flame optimization with deep neural network algorithm for e-healthcare systems,” Scientific Programming, vol. 2022, art. 4100352, pp. 1–12 (2022).[5] P. R. Kumar and B. Shilpa, An IoT-based smart healthcare system with edge intelligence computing, Apple Academic Press, pp. 31–46 (2024).[6] Sonam and Jyoti, “An Apriori algorithm-based framework for measuring and improving transactional dataset efficiency,” Journal of Statistics and Management Systems, vol. 28, no. 6, pp. 1179–1189 (2025).[7] V. R. R. Chirra, S. R. Uyyala, and V. K. K. Kolli, “Deep CNN: A machine learning approach for driver drowsiness detection based on eye state,” Revue d’Intelligence Artificielle, vol. 33, no. 6, pp. 461–466 (2019).[8] B. Shilpa, P. R. Kumar, and R. K. Jha, “Spreading factor optimization for interference mitigation in dense indoor LoRa networks,” in 2023 IEEE IAS Global Conf. Emerging Technologies (GlobConET), pp. 1–5 (May 2023).[9] J. Cockburn, M. Bartlett, J. Tanaka, J. Movellan, M. Pierce, and R. Schultz, “SmileMaze: A tutoring system in real-time facial expression perception and production in children with autism spectrum disorder,” in Proc. Workshop Facial Bodily Expressions Control Adaptation Games, Art. no. 3 (2008).[10] C. V. Rami Reddy, K. V. K. Kishore, D. Bhattacharyya, and T. H. Kim, “Multi-feature fusion based facial expression classification using DLBP and DCT,” Int. J. Softw. Eng. Its Appl., vol. 8, no. 9, pp. 55–68 (2014).[11] C. V. Ramireddy and K. V. K. Kishore, “Facial expression classification using Kernel based PCA with fused DCT and GWT features,” in 2013 IEEE Int. Conf. Comput. Intell. Comput. Res., Enathi, pp. 1–6 (2013).[12] P. R. Kumar, “Wireless mobile charger using inductive coupling,” J. Emerg. Technol. Innov. Res. (JETIR), vol. 5, no. 10, pp. 40–44 (2018).[13] S. Arvind, B. Unhelkar, S. S. Shankar, P. Chakrabarti, and S. V. Devika, “Advancing cyber threat detection through deep learning in management information systems,” Journal of Discrete Mathematical Sciences and Cryptography, vol. 27, no. 8, pp. 2409–2417 (2024).[14] V. Tumen, O. F. Soylemez, and B. Ergen, “Facial emotion recognition on a dataset using convolutional neural network,” in 2017 Int. Artif. Intell. Data Process. Symp. (IDAP), Malatya, pp. 1–5 (2017).[15] S. Sharma, A. K. Patel, and N. Menaria, “Integral transform of product of generalized k-Bessel function and generalized Hurwitz-Lerch Zeta function,” Journal of Interdisciplinary Mathematics, vol. 27, no. 8, pp. 1757–1764 (2024).
Views: 113Downloads: 11Citations: 0
