Multimodal Daily-Life Emotional Recognition Using Heart Rate and Speech Data From Wearables DOI Creative Commons

Eesun Moon,

A. S. M. Sharifuzzaman Sagar, Hyung Seok Kim

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 96635 - 96648

Published: Jan. 1, 2024

Human emotion plays a significant role in mental well-being, and recognizing these emotions daily life is essential. With the advancement of artificial intelligence, affective computing has paved way for effective applications enhancing emotional states everyday life. In practical daily-life scenarios, data sources that can be collected through simple low-cost wearables contribute to routines. Heart rate speech easily from affordable smartwatches without any other human intervention. data, directly correlated with physiological response, its rich expressiveness, together yield robust indicator human's condition. We conduct multimodal recognition (MER), integrating heart rate-based (HER) speech-based (SER) score-based fusion method. Our proposed MER achieves an overall accuracy 84.22%, surpassing single-modality models accuracies 57.65% HER 80.38% SER. The findings highlight practicality utilizing emotion-related conveniently smartwatches, thereby tracking accessibility scenarios. Furthermore, modalities proves more capturing than using single modality. Moreover, our system's lightweight architecture facilitates easy expansion incorporate additional modalities, ensuring durable precision even when not all are sensed, making it versatile pragmatic.

Language: Английский

Computationally efficient optimal control analysis for the mathematical model of Coronavirus pandemic DOI
Azhar Iqbal Kashif Butt, W. Ahmad, Muhammad Rafiq

et al.

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 234, P. 121094 - 121094

Published: Aug. 6, 2023

Language: Английский

Citations

24
Artificial Intelligence Techniques for Bankruptcy Prediction of Tunisian Companies: An Application of Machine Learning and Deep Learning-Based Models DOI Open Access

Manel Hamdi,

Sami Mestiri, Adnène Arbi

et al.

Journal of risk and financial management, Journal Year: 2024, Volume and Issue: 17(4), P. 132 - 132

Published: March 22, 2024

The present paper aims to compare the predictive performance of five models namely Linear Discriminant Analysis (LDA), Logistic Regression (LR), Decision Trees (DT), Support Vector Machine (SVM) and Random Forest (RF) forecast bankruptcy Tunisian companies. A Deep Neural Network (DNN) model is also applied conduct a prediction comparison with other statistical machine learning algorithms. data used for this empirical investigation covers 25 financial ratios large sample 732 companies from 2011–2017. To interpret results, three measures have been employed; accuracy percentage, F1 score, Area Under Curve (AUC). In conclusion, DNN shows higher in predicting compared conventional models, whereas random forest performs better than methods.

Language: Английский

Citations

9
Time Series Prediction in Industry 4.0: A Comprehensive Review and Prospects for Future Advancements DOI Creative Commons
Nataliia Kashpruk, C. Piskor-Ignatowicz, Jerzy Baranowski

et al.

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(22), P. 12374 - 12374

Published: Nov. 15, 2023

Time series prediction stands at the forefront of fourth industrial revolution (Industry 4.0), offering a crucial analytical tool for vast data streams generated by modern processes. This literature review systematically consolidates existing research on predictive analysis time within framework Industry 4.0, illustrating its critical role in enhancing operational foresight and strategic planning. Tracing evolution from first to revolution, paper delineates how each phase has incrementally set stage today’s data-centric manufacturing paradigms. It critically examines emergent technologies such as Internet things (IoT), artificial intelligence (AI), cloud computing, big analytics converge context 4.0 transform into actionable insights. Specifically, explores applications maintenance, production optimization, sales forecasting, anomaly detection, underscoring transformative impact accurate forecasting operations. The culminates call action dissemination management these technologies, proposing pathway leveraging drive societal economic advancement. Serving foundational compendium, this article aims inform guide ongoing practice intersection 4.0.

Language: Английский

Citations

18
Numerical solution of coupled system of Emden-Fowler equations using artificial neural network technique DOI Creative Commons
Ashish Kumar, M Suresh Kumar, Pranay Goswami

et al.

An International Journal of Optimization and Control Theories & Applications (IJOCTA), Journal Year: 2024, Volume and Issue: 14(1), P. 62 - 73

Published: Jan. 9, 2024

In this paper, a deep artificial neural network technique is proposed to solve the coupled system of Emden-Fowler equations. A vectorized form algorithm developed. Implementation and simulation performed using Python code. This implemented in various numerical examples, simulations are conducted. We have shown graphically how accurately method works. comparison solution exact error tables. also conducted comparative analysis our with alternative methods, including Bernstein collocation Homotopy method. The results presented efficiency accuracy demonstrated by these graphs

Language: Английский

Citations

6
Predicting U.S. bank failures and stress testing with machine learning algorithms DOI

W. C. Hu,

Chunli Shao, W. Zhang

et al.

Finance research letters, Journal Year: 2025, Volume and Issue: unknown, P. 106802 - 106802

Published: Jan. 1, 2025

Language: Английский

Citations

0
Numerical Analysis and Artificial Neural Networks for Solving Nonlinear Tuberculosis Model in SEITR Framework DOI Open Access

N. Jeeva,

K. M. Dharmalingam

Advanced Theory and Simulations, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 16, 2025

Abstract This study investigates an epidemiological model of tuberculosis dynamics by classifying the total population into five distinct compartments: susceptible, exposed, infected, treated, and recovered. To solve system nonlinear differential equations obtain approximate solutions for model, three analytical methods are utilized: transcendental‐exponential type proposed method (PNM), Homotopy perturbation (HPM), higher‐order inverse polynomial (HOIPM). Additionally, examines stochastic performance artificial neural networks trained using Levenberg–Marquardt algorithm (ANNs‐LMB) to offer a comprehensive evaluation model. The predictions generated ANNs‐LMB provide valuable benefits researchers, significantly improving their understanding infectious dynamics. Furthermore, error estimations demonstrate that PNM, HOIPM, highly effective in generating accurate solutions, closely matching those obtained from Runge–Kutta solver, surpassing HPM. These exhibit strong reliability efficiency, making them innovative tools addressing models simulating challenges. Moreover, analysis key parameters, including contact rate, infection tuberculosis‐related mortality reinfection treatment provides crucial insights model's behavior dynamics, paving way future research intervention strategies.

Language: Английский

Citations

0
A Swarm Optimized ANN-based Numerical Treatment of Nonlinear SEIR System based on Zika Virus DOI Open Access
Farhad Muhammad Riaz, Jawaria Ali Khan

Journal of Polytechnic, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 1

Published: Feb. 17, 2025

The purpose of the current study is to present numerical treatment a nonlinear mathematical SEIR model based on Zika virus using Mexican Hat Wavelet-based feed-forward artificial neural network (MHW-ANN) together with optimization scheme global search, Particle Swarm Optimization (PSO) and local search Sequential Quadratic Programming (SQP), i.e. MHW-ANN-PSO-SQP. an epidemic disease that can spread through transmission known as Aedes, its susceptible-exposed-infected-recovered, investigated dynamics spread. To solve error-based fitness function optimized hybrid computing validate precision, accuracy, stability, reliability, computational complexity designed framework various cases have been taken for virus. results obtained from MHW-ANN-PSO-SQP are compared well-known RK solver ANN-based (GA-ASA) confirm accuracy. At same time, absolute error validated precision scheme. Additionally, statistical analysis different operators performed convergence, reliability Furthermore, presented analyzed Mean Execution Time (MET).

Language: Английский

Citations

0
Model-free Adaptive Iterative Learning Control for Nonlinear Systems With Time-varying Constraints DOI
Wei Yan, Yongqi Zhang, Zi-Yuan Dong

et al.

International Journal of Control Automation and Systems, Journal Year: 2025, Volume and Issue: 23(3), P. 935 - 944

Published: March 10, 2025

Language: Английский

Citations

0
Modeling and analyzing the dynamics of brucellosis disease with vaccination in the fractional derivative under real cases DOI Creative Commons
Bashir Al‐Hdaibat, Muhammad Altaf Khan, Irfan Ahmad

et al.

Journal of Applied Mathematics and Computing, Journal Year: 2025, Volume and Issue: unknown

Published: March 21, 2025

Language: Английский

Citations

0
Accessibility of the three-year comprehensive prevention and control of brucellosis in Ningxia: a mathematical modeling study DOI Creative Commons
Wei Gong, Peng Sun,

Changsheng Zhai

et al.

BMC Infectious Diseases, Journal Year: 2023, Volume and Issue: 23(1)

Published: May 5, 2023

Brucellosis is a chronic zoonotic disease, and Ningxia one of the high prevalence regions in China. To mitigate spread brucellosis, government has implemented comprehensive prevention control plan (2022-2024). It meaningful to quantitatively evaluate accessibility this strategy.Based on transmission characteristics brucellosis Ningxia, we propose dynamical model sheep-human-environment, which coupling with stage structure sheep indirect environmental transmission. We first calculate basic reproduction number [Formula: see text] use fit data human brucellosis. Then, three widely applied strategies that is, slaughtering sicked sheep, health education risk practitioners, immunization adult are evaluated.The calculated as text], indicating will persist. The good alignment data. quantitative evaluation results show current strategy may not reach goal time. "Ningxia Prevention Control Special Three-Year Action Implementation Plan (2022-2024)" be achieved 2024 when increasing rate by 30[Formula: reduce 50[Formula: an increase 40[Formula: text].The demonstrate measures most effective for control, it necessary further strengthen multi-sectoral joint mechanism adopt integrated These can provide reliable basis optimizing Ningxia.

Language: Английский

Citations

9