A Channel Selection Method for Emotion Recognition From EEG Based on Swarm-Intelligence Algorithms

IEEE Access, Год журнала: 2021, Номер 9, С. 109889 - 109902

Опубликована: Янв. 1, 2021

Increasing demand for human-computer interaction applications has escalated the need automatic emotion recognition as emotions are essential natural communication. There various information sources that can be used recognizing emotions, such speech, facial expressions, body movements, and physiological signals. Among those signals more reliable better affective communication with machines since they almost impossible to control. Therefore, from EEG been a topic intensely investigated. Emotions experiences arise cognitive functions observed in different frequency bands involving multiple brain areas high accuracies is only possible large number of features extracted whole bands. Emotion regulation also requires integration thus functional connectivity between regions should considered. In this paper, we extract 736 based on spectral power phase-locking values. We particularly focus finding salient using swarm-intelligence (SI) algorithms. applied well-known classification algorithms positive negative feature sets selected by these Besides, all them commonly new set. report 56.27% 60.29% average; noting decreasing size 87.17% (from 94.40) an average accuracy 60.01± 8.93 was obtained random forest classifier. highlight efficient electrode locations recognition. As result, define 11 channels dominant promising results obtained.

Язык: Английский

---

Expert system for the parabolic trough collector control through classical and conformable transfer functions in ANNi-PSO

Expert Systems with Applications, Год журнала: 2025, Номер 280, С. 127343 - 127343

Опубликована: Апрель 9, 2025

Язык: Английский

---

Machine Learning Techniques for Renewable Energy Forecasting: A Comprehensive Review

Green energy and technology, Год журнала: 2022, Номер unknown, С. 3 - 39

Опубликована: Янв. 1, 2022

Язык: Английский

---

Thermo-economic optimization for the advanced material selection of fins and heat sinks

Chemical Engineering and Processing - Process Intensification, Год журнала: 2024, Номер unknown, С. 110109 - 110109

Опубликована: Дек. 1, 2024

Язык: Английский

---

Numerical investigation of the thermo-hydraulic performance of DNA inspired double and triple helix wire coils

Chemical Engineering and Processing - Process Intensification, Год журнала: 2021, Номер 171, С. 108736 - 108736

Опубликована: Ноя. 30, 2021

Язык: Английский

---

Neuro-evolutionary for time series forecasting and its application in hourly energy consumption prediction

Neural Computing and Applications, Год журнала: 2023, Номер 35(29), С. 21697 - 21707

Опубликована: Авг. 19, 2023

Язык: Английский

---

GIS-Based Evaluation of Disaster and Emergency Assembly Areas with AHP in Local Scale

Özgür Yayınları eBooks, Год журнала: 2023, Номер unknown

Опубликована: Июнь 22, 2023

In an area that is tectonically active and has various fault zones, earthquakes of varying magnitudes have occurred in our country both the past today. Like rest world, analyses research are being conducted to eliminate or minimize loss life property caused by earthquakes. this study, locations existing disaster emergency assembly areas Ovacık district were determined visualized on Google Earth. Furthermore, using Geographic Information Systems (GIS)-based analysis weighted overlay method, a risk classification was conducted. To end, basic inputs GIS environment, including slope, geology, distance rivers, roads, settlements, elevation, lines, land use, population data layers, used. Existing collection digitized incorporated into framework as vector layers. The sensitivity map study divided 3 classes. According map; regions high, middle low obtained 8.35%, 76.23% 15.42%, respectively. It observed within high zone. two geologically located Karabük formation (Teka). Because it contains coal levels, require more detailed field studies taking measures. Apart from this, been comply with other parameters

Язык: Английский

---

A New Co-Optimized Hybrid Model Based on Multi-Objective Optimization for Probabilistic Wind Power Forecasting in a Spatio–Temporal Framework

IEEE Access, Год журнала: 2023, Номер 11, С. 84885 - 84899

Опубликована: Янв. 1, 2023

Wind power generation is characterized by high intermittency and volatility owing to the stochastic nature of wind. In addition forecasting accuracy, uncertainty quantification can have a major impact on system energy management operations planning. this paper, new fully co-optimized hybrid short-term probabilistic Power Forecasting (WPF) model proposed for construction Prediction Intervals (PIs) in spatiotemporal framework. A Multi-Objective Improved Adaptive Particle Swarm Optimization (MOIAPSO) algorithm developed optimize model's parameters. PIs are generated nonlinear autoregressive networks with exogenous inputs (NARX) using Lower Upper Bound Estimation (LUBE) method. Unlike previous related work, components NARX-LUBE-MOIAPSO model, as well initial settings configuration parameters, determined based full co-optimization approach. The performed from quality training time trade-off perspective. Furthermore, framework introduced improve performance comprehension regional dynamics. comprises novel conditional methodology modeling dependencies binary Probabilistic Error (PFE) metric. tested publicly available datasets consisting turbine-specific measurements generate accurate forecasts efficient quantification, while maintaining computational complexity at relatively low levels compared other state-of-the-art WPF models.

Язык: Английский

---

An Intelligent Hybrid Model Using Artificial Neural Networks and Particle Swarm Optimization Technique For Financial Crisis Prediction

Pakistan Journal of Statistics and Operation Research, Год журнала: 2022, Номер unknown, С. 1015 - 1025

Опубликована: Дек. 6, 2022

Financial crisis prediction is a critical issue in the economic phenomenon. Correct predictions can provide knowledge for stakeholders to make policies preserve and increase stability. Several approaches predicting financial have been developed. However, classification model's performance accuracy, as well legal data, are insufficient usage real applications. So that, an efficient model required higher results. This paper adopts novel two-hybrid intelligent using Artificial Neural Network (ANN) Particle Swarm Optimization (PSO) optimization. At first, PSO technique produces hyperparameter value ANN fit best architecture. They weights thresholds. Then, they used predict of given dataset. In end, ANN-PSO generates conditions. The proposed implemented on time series data conditions Indonesia. Dataset was obtained from International Monetary Fund Indonesian Economic Statistics. Independent variable 13 potential indicators, namely imports, exports, trade exchange rates, foreign reserves, composite stock price index, deposit bank deposits, loan interest difference between BI rate FED rate, M1, M2 multiplier, ratio reserves. Meanwhile, dependent uses perfect signal based Pressure Index. A detailed statistical analysis dataset also by threshold convey Experimental shows that reliable different evaluation criteria. case studies show result predictive basically consistent with actual situation, which has greatly helped crisis.

Язык: Английский

---

Reflection of Industry 4.0 on Logistics Activities: Logistics 4.0

Özgür Yayınları eBooks, Год журнала: 2023, Номер unknown

Опубликована: Июнь 22, 2023

The idea of "Industry 4.0" was developed as a result the industrial industry's transition to digital manufacturing. term refers new era characterized by combination automation production processes, data exchange, artificial intelligence and internet. key goals Industry 4.0 are make processes more efficient, offer flexibility, develop smarter manufacturing environment. advancements brought about have had an impact on logistics sector, along with many other industries, highlighting Logistics 4.0. transformation in supply chain management. This concept aims from consumer flexible connected. encompasses number developments involving use technological innovations platforms. In this study, historical development discussed components examined detail. addition, information 4.0, which has effect is given. sample applications made activities within scope presented.

Язык: Английский

---