Adventures in data analysis: a systematic review of Deep Learning techniques for pattern recognition in cyber-physical-social systems

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 83(8), P. 22909 - 22973

Published: Aug. 9, 2023

Language: Английский

---

Factors Affecting Public Transportation Use during Pandemic

Tehnički glasnik, Journal Year: 2024, Volume and Issue: 18(3), P. 342 - 353

Published: July 15, 2024

For preventing the spread of COVID-19, health authorities urgently turned their attention to urban public transportation. It is believed that virus transmission more likely occur in transportation due increased exposure infected individuals closed and crowded spaces transport. This study aimed model effective factors use systems during a pandemic based on technology acceptance (TAM) theory planned behaviour (TPB). The methodology used was structural equation modeling, with 358 Iranian passengers Tehran participating data collected through questionnaire. underwent analysis by means partial least squares method assistance SMARTPLS software. results indicate passenger satisfaction affected positively significantly expectation service quality. Behavioral control, subjective norm, attitude, perceived usefulness (PU), ease (PEU) each contribute formation intention. Service quality, PU, PEU affect attitude. Finally, expectation, intention, system. Therefore, it can be inferred amalgamating TPB TAM serve as robust indicator passengers' inclination towards using situations, well actual usage it.

Language: Английский

---

A new QoS-aware service discovery technique in the Internet of Things using whale optimization and genetic algorithms

Journal of Engineering and Applied Science, Journal Year: 2024, Volume and Issue: 71(1)

Published: Jan. 3, 2024

Abstract Rapid technological advances have made daily life easier and more convenient in recent years. As an emerging technology, the Internet of Things (IoT) facilitates interactions between physical devices. With advent sensors features on everyday items, they become intelligent entities able to perform multiple functions as services. IoT enables routine activities intelligent, deeper communication, processes efficient. In dynamic landscape IoT, effective service discovery is key optimizing user experiences. A Quality Service (QoS)-aware technique proposed this paper address challenge. Through whale optimization genetic algorithms, our method aims streamline decision-making selection. The bio-inspired techniques employed approach facilitate services efficiently than traditional methods. Our results demonstrate superior performance regarding reduced data access time, optimized energy utilization, cost-effectiveness through comprehensive simulations.

Language: Английский

---

Optimal Band Selection Using Evolutionary Machine Learning to Improve the Accuracy of Hyper-spectral Images Classification: a Novel Migration-Based Particle Swarm Optimization

Journal of Classification, Journal Year: 2023, Volume and Issue: 40(3), P. 552 - 587

Published: Sept. 16, 2023

Language: Английский

---

Optimizing Long Short-Term Memory Network for Air Pollution Prediction Using a Novel Binary Chimp Optimization Algorithm

Electronics, Journal Year: 2023, Volume and Issue: 12(18), P. 3985 - 3985

Published: Sept. 21, 2023

Elevated levels of fine particulate matter (PM2.5) in the atmosphere present substantial risks to human health and welfare. The accurate assessment PM2.5 concentrations plays a pivotal role facilitating prompt responses by pertinent regulatory bodies mitigate air pollution. Additionally, it furnishes indispensable information for epidemiological studies concentrating on exposure. In recent years, predictive models based deep learning (DL) have offered promise improving accuracy efficiency quality forecasts when compared other approaches. Long short-term memory (LSTM) networks proven be effective time series forecasting tasks, including pollution prediction. However, optimizing LSTM enhanced remains an ongoing research area. this paper, we propose novel approach that integrates binary chimp optimization algorithm (BChOA) with optimize prediction models. proposed BChOA, inspired social behavior chimpanzees, provides powerful technique fine-tune architecture its parameters. evaluation results is performed using cross-validation methods such as coefficient determination (R2), accuracy, root mean square error (RMSE), receiver operating characteristic (ROC) curve. performance BChOA-LSTM model against eight DL architectures. Experimental evaluations real-world data demonstrate superior BChOA-based traditional algorithms. achieved highest 96.41% validation datasets, making most successful approach. show performs better than architectures terms R2 convergence curve, RMSE, accuracy.

Language: Английский

---

Digital Twin technology for multimodal-based smart mobility using hybrid Co-ABC optimization based deep CNN

Cluster Computing, Journal Year: 2025, Volume and Issue: 28(3)

Published: Jan. 21, 2025

Language: Английский

---

Anti-Attack Intrusion Detection Model Based on MPNN and Traffic Spatiotemporal Characteristics

Journal of Grid Computing, Journal Year: 2023, Volume and Issue: 21(4)

Published: Oct. 28, 2023

Language: Английский

---

Vehicle type classification in intelligent transportation systems using deep learning

Journal of Intelligent & Fuzzy Systems, Journal Year: 2024, Volume and Issue: 46(2), P. 5021 - 5032

Published: Jan. 9, 2024

Intelligent Transportation Systems (ITS) have experienced significant growth over the past decade thanks to advances in control, communication, and information technology applied vehicles, roads, traffic control systems. Vehicle type classification plays a vital role implementing ITS because of its ability collect useful information, enable future development transport infrastructures, increase human comfort. As branch machine learning, deep learning represents frontier for artificial intelligence, which seeks be closer primary goal. Deep is powerful tool classifying vehicle types it can capture complex data characteristics learn from large amounts data. This means that used accurately classify generate valuable insights improve management. Researchers successfully adopted these algorithms as solution propose optimal vehicle-type strategies. paper highlights solving problem, reviewing state-of-the-art approaches this field.

Language: Английский

---

Multi-Agent Systems for Collaborative Inference Based on Deep Policy Q-Inference Network

Journal of Grid Computing, Journal Year: 2024, Volume and Issue: 22(1)

Published: Feb. 29, 2024

Language: Английский

---

DeepShield: A Hybrid Deep Learning Approach for Effective Network Intrusion Detection

International Journal of Advanced Computer Science and Applications, Journal Year: 2023, Volume and Issue: 14(7)

Published: Jan. 1, 2023

In today's rapidly evolving digital landscape, ensuring the security of networks and systems has become more crucial than ever before. The ever-present threat hackers intruders attempting to disrupt compromise online services highlights pressing need for robust measures. With continuous advancement systems, new dangers arise, but so do innovative solutions. One such solution is implementation Network Intrusion Detection Systems (NIDSs), which play a pivotal role in identifying potential threats computer by categorizing network traffic. However, effectiveness an intrusion detection system lies its ability prepare data identify critical attributes necessary constructing classifiers. light this, this paper proposes, DeepShield, cutting-edge NIDS that harnesses power deep learning leverages hybrid feature selection approach optimal performance. DeepShield consists three essential steps: selection, rule assessment, detection. By combining strengths machine technologies, developed excels detecting intrusions. process begins capturing packets from network, are then carefully preprocessed reduce their size while retaining information. These refined fed into algorithm, employs characteristics learn test patterns. Simulation results demonstrate superiority over previous approaches. achieves exceptional level accuracy malicious attacks, as evidenced outstanding performance on widely recognized CSE-CIC-DS2018 dataset.

Language: Английский

---