Article
Computer Science, Information Systems
Madhu Sudan Kumar, Anubhav Choudhary, Indrajeet Gupta, Prasanta K. Jana
Summary: This paper proposes an effective two-phase algorithm for provisioning cloud resources for workflow applications, taking into account the type of tasks and constraints of the cloud model to minimize makespan and resource wastage. The superiority of the proposed approach is validated through simulation results using five benchmark scientific workflows.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Hardware & Architecture
Ming Li, Zhong Fang, Wanwan Cao, Yong Ma, Shang Wu, Yang Guo, Yu Xue, Romany F. Mansour
Summary: This study proposes an improved random forest residential electricity classification method based on cloud computing, which optimizes the internal parameters of the random forest algorithm and incorporates the Drosophila algorithm to achieve more accurate electricity consumption classification. The effectiveness of the model is verified through experiments, demonstrating the feasibility of the proposed method.
COMPUTER SYSTEMS SCIENCE AND ENGINEERING
(2021)
Article
Computer Science, Information Systems
Tarun Ganesh Palla, Shahab Tayeb
Summary: The research proposed a novel approach using machine learning algorithm to detect Mirai malware, which achieved satisfactory performance in the experiment and conducted comparative analysis with random forest model.
Article
Automation & Control Systems
Runze Gao, Qiwen Li, Li Dai, Yufeng Zhan, Yuanqing Xia
Summary: This paper proposes a workflow-based fast data-driven predictive control (DPC) method in cloud-edge collaborative architecture to improve computational efficiency. A cloud-edge collaborative scheme is designed to tackle uncertainty in cloud workflow processing, and an autonomous cloud control experimental system is implemented to execute the DPC controller. Evaluations demonstrate significant reductions in computation time for real-time control examples.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Chemistry, Analytical
Omer Kaspi, Olga Girshevitz, Hanoch Senderowitz
Summary: This paper presents a structured workflow for glass fragment analysis based on a combination of Elemental Analysis using PIXE and Machine Learning tools, achieving high accuracy in identifying the origins of glass fragments and demonstrating its applicability in real-life forensic events. The standard, reproducible methodology proposed in this study can be utilized in various forensic domains beyond glass fragments.
Article
Computer Science, Information Systems
Puneet Sharma, Manoj Kumar, Ashish Sharma
Summary: With the increasing number of users in social media, the handling of files has also increased. To manage this load, service providers are using cloud servers. This paper proposes a hybrid algorithm that combines the features of Random Forest with AdaBoost to address the difficulty of file identification and clustering. The algorithm, called Internet of Thing (IoT) data file formatting (IDFF), classifies data as text, image, audio, and video, and achieves better accuracy.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Arijit Roy, Sudip Misra, Prerona Dutta
Summary: The presence of dumb nodes in the sensor-cloud environment can degrade system performance. This article proposes a dynamic pricing scheme, DISCLOUD, which considers the existence of dumb nodes and aims to maximize the profit of the Sensor-Cloud Service Provider (SCSP) while ensuring Quality of Service (QoS) for end-users.
IEEE TRANSACTIONS ON CLOUD COMPUTING
(2022)
Article
Computer Science, Hardware & Architecture
Ali Banaei, Mohsen Sharifi
Summary: This paper introduces ETAS, a predictive scheduling scheme for function scheduling within worker nodes of the Apache OpenWhisk serverless platform, which aims to reduce response times of invocations by utilizing factors such as execution times, arrival times, and containers' status for scheduling. The implementation of ETAS in Apache OpenWhisk shows a 30% reduction in average waiting time and a 40% increase in throughput compared to other scheduling schemes.
JOURNAL OF SUPERCOMPUTING
(2022)
Article
Computer Science, Hardware & Architecture
Zain Ulabedin, Babar Nazir
Summary: This paper introduces a workflow scheduling technique to overcome data transfer and execute workflow tasks within deadline and budget constraints. By clustering and distributing data, as well as utilizing replication-based partial critical path (R-PCP) technique to schedule tasks, the strategy efficiently reduces data movement and executes all chosen workflows within user specified budget and deadline.
JOURNAL OF SUPERCOMPUTING
(2021)
Article
Computer Science, Hardware & Architecture
Anurina Tarafdar, Kamalesh Karmakar, Rajib K. Das, Sunirmal Khatua
Summary: This paper proposes a novel workflow scheduling approach for the Workflow as a Service (WaaS) platform, which reduces the average makespan of the workflows, improves energy efficiency, and reduces the resource renting cost of the Cloud resources. The scheduling model includes containers, virtual machines (VMs), and hosts, and a suitable scaling policy is proposed to improve resource utilization. Extensive simulations with real-world workflows and comparison with state-of-the-art algorithms demonstrate the efficacy of the approach in improving performance, energy efficiency, and reducing monetary cost.
COMPUTERS & ELECTRICAL ENGINEERING
(2023)
Article
Engineering, Chemical
Pablo Diaz, Juan C. Salas, Aldo Cipriano, Felipe Nunez
Summary: As mineral processing operations become more complex, automation and control are crucial. In modern mineral processing, utilizing machine-learning models and predictive control strategies can enhance the efficiency of paste thickening processes.
MINERALS ENGINEERING
(2021)
Article
Statistics & Probability
Torsten Hothorn, Achim Zeileis
Summary: This article discusses regression models for supervised learning problems with continuous response, suggesting a more general understanding of regression models as models for conditional distributions. Quantile regression forests are highlighted among algorithms estimating conditional distributions. A novel approach based on a parametric family of distributions characterized by their transformation function is proposed, along with a dedicated transformation tree algorithm for detecting distributional changes. Prediction intervals and inference procedures are provided by the resulting predictive distributions, making them fully parametric yet very general.
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
(2021)
Article
Multidisciplinary Sciences
Xiaofei Sun, Jingyuan Zhu, Bin Chen, Hengzhi You, Huiqing Xu
Summary: In this study, a two-stage transfer learning workflow was proposed to develop a novel prediction model for accurately predicting drug activity and toxicity of targets with insufficient observations. By building a balanced dataset and utilizing transfer learning strategies, the accuracy of the models in understudied targets was improved.
Article
Engineering, Chemical
Heng Liu, Hao Zheng, Zhenhe Jia, Binghui Zhou, Yan Liu, Xuelu Chen, Yajun Feng, Li Wei, Weijie Yang, Hao Li
Summary: This article introduces CatMath, an online platform for generating catalytic volcano activity models, and provides functions for analyzing catalyst activity and electrocatalytic surface states.
FRONTIERS OF CHEMICAL SCIENCE AND ENGINEERING
(2023)
Article
Environmental Sciences
Lingfeng Liao, Shengjun Tang, Jianghai Liao, Xiaoming Li, Weixi Wang, Yaxin Li, Renzhong Guo
Summary: This paper proposes a robust and effective point cloud classification method that integrates point cloud supervoxels and their locally convex connected patches into a random forest classifier. It improves the accuracy of point cloud feature calculation and reduces computational cost. Two benchmark tests show that the proposed method achieves state-of-the-art performance and successfully classifies point clouds with great variation in challenging scenes.
Editorial Material
Biochemistry & Molecular Biology
Antreas Afantitis, Georgia Melagraki
CURRENT MEDICINAL CHEMISTRY
(2020)
Article
Biochemistry & Molecular Biology
Christiana Magkrioti, Eleanna Kaffe, Elli-Anna Stylianaki, Camelia Sidahmet, Georgia Melagraki, Antreas Afantitis, Alexios N. Matralis, Vassilis Aidinis
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2020)
Editorial Material
Chemistry, Multidisciplinary
Iseult Lynch, Antreas Afantitis, Dario Greco, Maria Dusinska, Miguel A. Banares, Georgia Melagraki
Review
Biochemistry & Molecular Biology
Varnavas D. Mouchlis, Antreas Afantitis, Angela Serra, Michele Fratello, Anastasios G. Papadiamantis, Vassilis Aidinis, Iseult Lynch, Dario Greco, Georgia Melagraki
Summary: De novo drug design is a process of generating novel molecular structures using computational methods, with traditional approaches including structure-based and ligand-based design. Artificial intelligence and machine learning have a positive impact in this field.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2021)
Article
Multidisciplinary Sciences
Laura Aliisa Saarimaki, Antonio Federico, Iseult Lynch, Anastasios G. Papadiamantis, Andreas Tsoumanis, Georgia Melagraki, Antreas Afantitis, Angela Serra, Dario Greco
Summary: Toxicogenomics is increasingly used to understand the toxicity mechanisms of engineered nanomaterials, but a unified collection of transcriptomics data for ENMs is currently lacking. A curated dataset from human, mouse, and rat exposures to ENMs has been compiled to improve accessibility and reusability of the data.
Article
Environmental Sciences
Anastasios G. Papadiamantis, Antreas Afantitis, Andreas Tsoumanis, Eugenia Valsami-Jones, Iseult Lynch, Georgia Melagraki
Summary: The physicochemical characterisation data of 69 engineered nanomaterials has been utilized to develop a nanoinformatics model for predicting the ENM zeta-potential. The model includes five critical parameters, such as ENM size and coating, as well as three molecular descriptors, each of which significantly influences the zeta-potential values. The model is available as a web service for the community through specific Horizon 2020 projects.
Review
Biochemistry & Molecular Biology
Mary Gulumian, Charlene Andraos, Antreas Afantitis, Tomasz Puzyn, Neil J. Coville
Summary: The physicochemical properties of nanomaterials have an impact on their toxicity and pathogenicity, with nanotopography being an important factor. Despite its significance, the role of surface topography in nanotoxicity is often overlooked. By manipulating surface topography and applying principles from catalysis, it is possible to create safer nanomaterials by reducing surface properties contributing to toxicity.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2021)
Article
Biochemistry & Molecular Biology
Dimitra Papadopoulou, Antonios Drakopoulos, Panagiotis Lagarias, Georgia Melagraki, George Kollias, Antreas Afantitis
Summary: The study identified potential anti-TNF small molecule inhibitors in vitro, with Nepalensinol B and Miyabenol A showing efficacy in reducing TNF-induced cytotoxicity. Nepalensinol B was also found to abolish TNF-TNFR1 binding at non-toxic concentrations.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2021)
Article
Chemistry, Medicinal
Dimitra-Danai Varsou, Nikoletta-Maria Koutroumpa, Haralambos Sarimveis
Summary: This study presents a computational workflow for grouping engineered nanomaterials (ENMs) and predicting their toxicity-related end points. A mixed integer-linear optimization program (MILP) problem is formulated to automatically filter noisy variables and develop specific predictive models for each group. The method demonstrates good performance through application to benchmark datasets and comparison with alternative predictive modeling approaches.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2021)
Review
Materials Science, Biomaterials
Zhiling Guo, Swaroop Chakraborty, Fazel Abdolahpur Monikh, Dimitra-Danai Varsou, Andrew J. Chetwynd, Antreas Afantitis, Iseult Lynch, Peng Zhang
Summary: The research reviews the impact of surface functionalization of graphene-based materials on nanotoxicity and safe design, including studies on intentionally designed functions for applications as well as unintentionally acquired effects from the environment and biota.
Article
Environmental Sciences
Dimitra-Danai Varsou, Laura-Jayne A. Ellis, Antreas Afantitis, Georgia Melagraki, Iseult Lynch
Summary: The study presents eco-toxicological read-across models for predicting the toxicity of differently aged nanomaterials on Daphnia magna, with the finding that presence of natural organic matter in the medium reduces the toxicity. Nanomaterials were grouped into freshly dispersed and 2-year-aged categories, with in silico analysis identifying key features driving toxicity in each group. The predictive models have been validated and are recommended for regulatory purposes.
Article
Chemistry, Multidisciplinary
P. Tsiros, N. Cheimarios, A. Tsoumanis, A. C. O. Jensen, G. Melagraki, I Lynch, H. Sarimveis, A. Afantitis
Summary: Integrated approaches to testing and assessment (IATA) combine different sources of information to characterize the hazards of chemicals, including nanomaterials. This study presents three computational approaches that can generate data relevant to human health risk assessment. By comparing the performance of these models under different conditions, their capabilities and potential contribution to a nanomaterial-specific IATA for occupational exposure can be evaluated.
ENVIRONMENTAL SCIENCE-NANO
(2022)
Article
Chemistry, Multidisciplinary
Jeaphianne van Rijn, Antreas Afantitis, Mustafa Culha, Maria Dusinska, Thomas E. Exner, Nina Jeliazkova, Eleonora Marta Longhin, Iseult Lynch, Georgia Melagraki, Penny Nymark, Anastasios G. Papadiamantis, David A. Winkler, Hulya Yilmaz, Egon Willighagen
Summary: Management of nanomaterials and nanosafety data requires a unique identifier for each nanomaterial, which is provided by the European Registry of Materials Identifier. This identifier ensures the linking of internal project documentation with publicly released data and knowledge for specific nanomaterials, and has been applied in H2020-funded nanosafety projects.
JOURNAL OF CHEMINFORMATICS
(2022)
Article
Chemistry, Medicinal
Dimitra-Danai Varsou, Haralambos Sarimveis
Summary: In this study, a computational methodology called deimos is presented for optimal grouping and prediction of toxicity-related properties of engineered nanomaterials (ENMs). The method utilizes a mixed-integer optimization program (MILP) to automatically select features, define grouping boundaries, and develop linear regression models in each group. Benchmark datasets and comparison with other predictive modeling approaches demonstrate the performance of deimos. Importantly, this method is not limited to ENMs toxicity prediction and can be applied to property prediction of other chemical entities.
MOLECULAR INFORMATICS
(2023)
Article
Plant Sciences
Peng Zhang, Zhiling Guo, Sami Ullah, Georgia Melagraki, Antreas Afantitis, Iseult Lynch
Summary: This Perspective discusses the applications of nanotechnology and artificial intelligence in agriculture, highlighting the opportunities and challenges of using these technologies to achieve sustainable and precision agriculture. By integrating knowledge and adopting new approaches, exciting opportunities can be created for sustainable food production.
