A5064344312- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- Privacy-Preserving Technologies in Data
- Parallel Computing and Optimization Techniques
- Adversarial Robustness in Machine Learning
- Network Security and Intrusion Detection
- Data Stream Mining Techniques
- Anomaly Detection Techniques and Applications
- Topic Modeling
- Cryptography and Data Security
- Complex Network Analysis Techniques
- Advanced Malware Detection Techniques
- Scientific Computing and Data Management
- Natural Language Processing Techniques
- Data Visualization and Analytics
- Metaheuristic Optimization Algorithms Research
- Optimization and Search Problems
- IoT and Edge/Fog Computing
- Advanced Text Analysis Techniques
- Meteorological Phenomena and Simulations
- Multimodal Machine Learning Applications
- Air Quality Monitoring and Forecasting
- Advanced Data Storage Technologies
- Artificial Intelligence in Healthcare and Education
- Mobile Crowdsensing and Crowdsourcing
Hanoi University of Science and Technology
2016-2025
Hung Yen University of Technology and Education
2025
Slovak Academy of Sciences
2015-2024
Institute of Informatics of the Slovak Academy of Sciences
2015-2024
Slovak University of Technology in Bratislava
2021-2024
University of Minnesota
2024
Seoul National University of Science and Technology
2024
RMIT Vietnam
2024
VinUniversity
2023
Macquarie University
2023
The combined impact of new computing resources and techniques with an increasing avalanche large datasets, is transforming many research areas may lead to technological breakthroughs that can be used by billions people. In the recent years, Machine Learning especially its subfield Deep have seen impressive advances. Techniques developed within these two fields are now able analyze learn from huge amounts real world examples in a disparate formats. While number algorithms extensive growing,...
In this paper we propose a distributed architecture to provide machine learning practitioners with set of tools and cloud services that cover the whole development cycle: ranging from models creation, training, validation testing serving as service, sharing publication. such respect, DEEP-Hybrid-DataCloud framework allows transparent access existing e-Infrastructures, effectively exploiting resources for most compute-intensive tasks coming cycle. Moreover, it provides scientists...
Time-ordered data are widely available in many real-life areas like traffic transportation, economic growth, weather prediction, as well monitoring and distributed system workloads more. Recently, deep learning models often applied to solve time-series prediction due their quality. While such recurrent neural networks the most well-known this direction, convolutional (CNNs) is more known for image processing. However, CNNs also a strong candidate sequence modeling forecasting. In general,...
The work presented in this paper deals with a proactive network monitoring for security and protection of computing infrastructures. We provide an exploitation intelligent module, the form as machine learning application using deep modeling, order to enhance functionality intrusion detection system supervising traffic flows. Currently, systems well near real-time they effectively deal threats reactive way. Deep is emerging generation artificial intelligence techniques one most promising...
Large language models have the potential to enhance equitable access health information, but their poor performance in some languages could exacerbate digital divide healthcare, say <b>Arthur Tang and colleagues</b>
Today, almost all clouds only offer auto-scaling functions using resource usage thresholds, which are defined by users. Meanwhile, applying prediction-based to still faces a problem of inaccurate forecast during operation in practice even though the deal with univariate monitoring data. Up until now, there very few efforts simultaneously process multiple metrics predict utilization. The motivation for this multivariate processing is that could be some correlations among and they have...
Botnets play major roles in a vast number of threats to network security, such as DDoS attacks, generation spam emails, information theft. Detecting is difficult task due the complexity and performance issues when analyzing huge amount data from real large-scale networks. In Botnet malware, use Domain Generation Algorithms allows decrease possibility be detected using white list - blacklist scheme thus DGA have higher survival. This paper proposes detection based on DNS traffic analysis...
The works presented in this paper addresses the robust population-based global optimization that is influenced by simplicity and efficiency principles introduced two new generation algorithms. Galactic Swarm Optimization inspired motion of stars, galaxies, superclusters galaxies under influence gravity. It acts well as a controller whole process employing multiple flexible cycles exploration exploitation phases to find new, better solutions. However, still suffers poverty phase, which...
A resolved CFD–DEM coupling model for the simulation of particulate flows is proposed in this work. The Volume Penalisation (VP) method, which a family continuous forcing Immersed Boundary (IB) employed to express particle–fluid interaction. smooth mask function used avoid sharp transition between solid (particle) and fluid domains that may cause numerical oscillation with moving particles. Optimal permeability reduce error associated VP method. It determined as only interface thickness...
Abstract The aim of the present study is to recover a waste titania catalyst from α‐pinene isomerization, recycle it and use in production inorganic pigments. acid‐contaminated (ACT), prepared ilmenite ore by sulphation, has potential function as solid acid due presence both Brønsted Lewis sites, with specific surface area 163 m 2 g −1 . catalytic efficiency was evaluated isomerization turpentine (83.8 wt.% α‐pinene) at reaction temperature 120°C, using turpentine:catalyst mass ratio 4:1....
The performance of hybrid satellite-terrestrial relaying (HSTR) networks is investigated in this work. Specifically, we examine the trade-off between reliability and security HSTR using two key parameters: outage probability (OP) intercept (IP). Both metrics are derived closed-form expressions under assumption imperfect channel state information (CSI) for legitimate channels. Additionally, a directional antenna employed to compensate significant path loss caused by long transmission distance...
Designing prediction-based auto-scaling systems for cloud computing is an attractive topic scientists today. However, there are many barriers, which must be solved before applying these to practice. Some challenges include: improving accuracy prediction models, finding a simple and effective forecast method instead of complex techniques, processing multivariate resource metrics at the same time. So far, no existing proactive solutions clouds that have addressed all those challenges. In this...
Time series data is widely accessible in many life areas like economy, weather, stock price, retail sales, distributed system workloads. While studies have focused on improving existing prediction techniques accuracy aspect, less efforts paid towards simple but efficient forecasting models order to keep the balance between computation cost and accuracy. In this work, we propose a novel time model, which aims both model simplicity The core of built based extreme learning machine. Due random...
Cloud computing has emerged as an optimal option for almost all computational problems today. Using cloud services, customers and providers come to terms of usage conditions defined in Service Agreement Layer (SLA), which specifies acceptable Quality (QoS) metric levels. From the view cloud-based software developers, their application-level SLA must be mapped provided virtual resource-level SLA. Hence, one important challenges clouds today is improve QoS resources. In this paper, we focus on...
Public administration frequently deals with geographically scattered personal data between multiple government locations and organizations. As digital technologies advance, public is increasingly relying on collaborative intelligence while protecting individual privacy. In this context, federated learning has become known as a potential technique to train machine models private distributed maintaining This work looks at the trade-off privacy assurances vulnerability membership inference...
Iron(II) tris-bipyridine, [FeII(bpy)3]2+, is a historically significant organometallic coordination complex with attractive redox and photophysical properties. With respect to energy storage, it low-cost, high-redox potential thus for use as catholyte in aqueous flow batteries. Despite these favorable characteristics, its oxidized Fe(III) form undergoes dimerization μ-O-[FeIII(bpy)2(H2O)]24+, leading dramatic ∼0.7 V decrease during battery discharge. To date, the energetics complete...
Massive data collected on public roads for autonomous driving has become more popular in many locations the world. More leads to concerns about privacy, including but not limited pedestrian faces and surrounding vehicle license plates, which urges robust solutions detecting anonymizing them realistic road-driving scenarios. Existing datasets both face plate detection are either focused or only parking lots. In this paper, we introduce a challenging dataset domain. The is aggregated from...
Five-second testing is a method commonly used by user research professionals to assess users' first impressions of interfaces or product designs. Its rule thumb, that five seconds generally the right amount time for users report realistic and relevant impressions, misrepresents reality human cognition. Users possess disparate levels cognitive ability processing stimuli varied visual complexity. We conducted complex experiment where participants complete an evaluation their – working memory...