Article
Chemistry, Multidisciplinary
Andrey Ferriyan, Achmad Husni Thamrin, Keiji Takeda, Jun Murai
Summary: This paper presents the HIKARI-2021 dataset, which contains encrypted synthetic attacks and benign traffic to address the lack of up-to-date datasets for evaluating intrusion detection systems. The dataset meets content and process requirements, and is made available to enable future dataset developments.
APPLIED SCIENCES-BASEL
(2021)
Article
Automation & Control Systems
Xiaokang Zhou, Yiyong Hu, Wei Liang, Jianhua Ma, Qun Jin
Summary: The article introduces a VLSTM model to address imbalance and high-dimension issues in industrial big data, which significantly improves accuracy and reduces false positives in anomaly detection.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Review
Computer Science, Information Systems
Ridhima Rani, Meenu Khurana, Ajay Kumar, Neeraj Kumar
Summary: This article highlights the importance of using dimensionality reduction techniques for processing high-dimensional data in the age of big data. It provides an overview of the advantages, properties, taxonomy, and evaluation parameters of dimensionality reduction techniques. The article also discusses the future research challenges and the applicability of dimensionality reduction in different domains, particularly in improving the storage and processing of big data in IoT applications.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Dania Herzalla, Willian Tessaro Lunardi, Martin Andreoni
Summary: This paper introduces the TII-SSRC-23 dataset, which is diverse and aligned with the contemporary network environment, providing an important tool for intrusion detection. Additionally, the study analyzes the importance of features and establishes benchmarks for intrusion detection methodologies through experiments.
Article
Computer Science, Information Systems
Myat Cho Mon Oo, Thandar Thein
Summary: In this paper, an efficient predictive analytics system for high dimensional big data is proposed by enhancing scalable random forest algorithm on the Apache Spark platform.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
T. M. Tariq Adnan, Md Mehrab Tanjim, Muhammad Abdullah Adnan
Summary: PCA is a popular technique for dimensionality reduction, but scalability issues arise in high dimensions. This study proposes a solution based on the zero-noise-limit Probabilistic PCA model, introducing a block-division method to suppress intermediate data explosion and achieve efficient communication in a geographically distributed environment.
INFORMATION SYSTEMS
(2021)
Article
Multidisciplinary Sciences
Zhenfei Wang, Yan Wang, Liying Zhang, Chuchu Zhang, Xingjin Zhang
Summary: The paper proposes an incremental high order singular value decomposition dimensionality reduction algorithm (icHOSVD) for unstructured campus big data, which combines tensioning of features and sub-tensor fusion to solve the representation and fusion issue of different data types, improving dimensionality reduction efficiency.
Article
Computer Science, Hardware & Architecture
Mehmet Bozdal, Kadir Ileri, Ali Ozkahraman
Summary: The Internet of Things (IoT) has transformed distributed cyber-physical systems, like city-wide water treatment systems, but also brought cybersecurity threats. This research proposes a new method using an enhanced 1D Convolutional Neural Network (CNN) model with a Gated Recurrent Unit (GRU) to secure the SWaT dataset. The proposed method outperforms existing methods with 99.68% accuracy and a 98.69% F1 score. The study also explores dimensionality reduction methods, and finds that PCA provides better performance compared to other techniques, achieving a 90.2% input dimension reduction with only a 2.8% and 2.6% decrease in accuracy and F1 score respectively. This research addresses the critical need for robust cybersecurity measures in IoT-enabled water treatment systems and considers the trade-off between dimensionality reduction and intrusion detection accuracy.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Biochemical Research Methods
Shuyi Zhang, Jacob R. Leistico, Raymond J. Cho, Jeffrey B. Cheng, Jun S. Song
Summary: This study introduces two spectral algorithms on multilayer graphs for clustering cells in multi-omic single-cell sequencing datasets, demonstrating the WLL method as a new spectral graph theoretic reformulation of the popular Seurat weighted nearest neighbor algorithm.
Article
Engineering, Industrial
Shigui Ma, Yong He, Ran Gu
Summary: This study examines the impact of competition and cooperation on low-carbon scenic spots and travel agencies within a low-carbon tourism supply chain network. The results indicate that tourist preferences for low-carbon products enhance the overall low-carbon level of tourism products, while competition between scenic spots leads to a decrease in low-carbon level. Additionally, the study suggests that cooperation among network members is not always beneficial for the performance of the entire network.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Hudhaifa Mohammed Abdulwahab, S. Ajitha, Mufeed Ahmed Naji Saif
Summary: This article provides a comprehensive review of the latest feature selection (FS) approaches in the context of big data. It categorizes the existing methods based on their nature, search strategy, evaluation process, and feature structure. The article presents qualitative and quantitative analyses of FS methods, as well as experimental comparisons to evaluate their performance. It also highlights the research issues and open challenges in FS, serving as a guide for future research directions.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Mozamel M. Saeed
Summary: This study aims to improve the performance of classifiers in identifying signatures of unknown attacks and establishes a hybrid classifier model based on the evaluation of commonly used classifiers. A quantitative methodology was adopted to collect and interpret data, and the evaluation was conducted in virtual networked environments with traffic workloads. The study reveals that certain features make significant contributions to anomaly detection.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Automation & Control Systems
Hassan Ismkhan, Mohammad Izadi
Summary: This article proposes a simple dimensionality reduction technique to speed up the process of finding the nearest neighbors in clustering algorithms. It is applied to accelerate the k-means algorithm and yields favorable results when compared with recent state-of-the-arts.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Kehua Guo, Yifei Wang, Jian Kang, Jian Zhang, Rui Cao
Summary: With the advancement of technology in the big data era, the development of intelligent medical diagnoses has been promoted by the increase in data in the medical field. Core dataset extraction from unlabeled medical big data can improve training efficiency and reduce human resource requirements.
Article
Computer Science, Information Systems
Waleed Albattah, Rehan Ullah Khan, Mohammed F. Alsharekh, Samer F. Khasawneh
Summary: This study aims to analyze a large amount of data with classical machine learning and investigates the impact of different random sampling techniques on model performance. Results suggest that using feature selection techniques and machine learning classifiers can enhance model performance when handling large datasets.
Article
Computer Science, Artificial Intelligence
Shashi Shekhar, Hitendra Garg, Rohit Agrawal, Shivendra Shivani, Bhisham Sharma
Summary: This paper describes a self-learning Hierarchical LSTM technique for classifying hatred and trolling contents in social media code-mixed data. The proposed method achieves high accuracy by mining the intention of the user for using hatred words in context.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Computer Science, Information Systems
Sheetal Garg, Rohit Ahuja, Raman Singh, Ivan Perl
Summary: This paper proposes a prediction model to forecast resource utilization of physical machines in order to reduce energy consumption in data centers. By processing the raw time series workload and employing clustering algorithms, the accuracy and effectiveness of the predictions are improved. The proposed model is evaluated using the Google cluster trace usage dataset.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2023)
Article
Environmental Sciences
Payal Mittal, Akashdeep Sharma, Raman Singh
Summary: Unmanned aerial vehicles (UAVs) equipped with mounted camera sensors are used for various remote sensing applications, such as search and rescue, estimation of endangered flora and fauna, and emergency responses. The proposed MLP Mixer architecture for classifying burned piles in dense forests outperforms transformer approaches in terms of performance, achieving a substantial accuracy of 77.23. The classification system was evaluated on a dataset of pile photos taken during a burning pile of debris in an Arizona pine forest.
JOURNAL OF APPLIED REMOTE SENSING
(2023)
Article
Energy & Fuels
Sanjiv Kumar Jain, Sandeep Bhongade, Shweta Agrawal, Abolfazl Mehbodniya, Bhisham Sharma, Subrata Chowdhury, Julian L. Webber
Summary: This study focuses on the load frequency control of a two-area thermal generation system with solar generation in one of the control areas. A novel grey wolf optimizer is used to tune the controller gains, and sensitivity analyses are conducted to account for deviations in time constants. The proposed algorithm shows good dynamic performance in terms of settling time, overshoot values, and undershoot values, with faster zero deviation achieved in solar-based cases compared to those without solar integration.
Article
Chemistry, Analytical
Saurabh Singhal, Senthil Athithan, Madani Abdu Alomar, Rakesh Kumar, Bhisham Sharma, Gautam Srivastava, Jerry Chun-Wei Lin
Summary: Data centers generate large amounts of data as cloud-based smart grids replace traditional grids. The increase in automated systems has led to the rise of cloud computing, which helps enterprises provide services at low cost and high efficiency. Despite challenges such as resource management, longer response and processing time, and increased energy consumption, cloud computing is being increasingly used. Fog computing, an extension of cloud computing, reduces traffic, improves security, and speeds up processes. Both cloud and fog computing contribute to energy savings in smart grids. The paper proposes a load-balancing approach using Rock Hyrax Optimization (RHO) in Smart Grids to optimize response time and energy consumption. The algorithm assigns tasks to virtual machines and shuts off unused ones, reducing energy consumption. The proposed method shows better and quicker response time, lower energy requirements, and improved performance compared to static and dynamic algorithms. It reduces processing time by 26%, response time by 15%, energy consumption by 29%, cost by 6%, and delay by 14%.
Article
Chemistry, Analytical
Balraj Kumar, Neeraj Sharma, Bhisham Sharma, Norbert Herencsar, Gautam Srivastava
Summary: Recommender systems are important but face challenges in producing quality recommendations due to sparsity issues. This study introduces a hierarchical Bayesian hybrid recommendation model, RCTR-SMF, for recommending music artists to users. The model integrates social matrix factorization and link probability functions into collaborative topic regression-based recommender systems, utilizing auxiliary domain knowledge for better prediction accuracy. The model addresses sparsity and cold-start problems, and shows superior performance compared to other state-of-the-art algorithms, as demonstrated on a large social media dataset.
Article
Computer Science, Information Systems
Saleem Raja Abdul Samad, Sundarvadivazhagan Balasubaramanian, Amna Salim Al-Kaabi, Bhisham Sharma, Subrata Chowdhury, Abolfazl Mehbodniya, Julian L. L. Webber, Ali Bostani
Summary: Phishing exploits the tendency of people to share personal information online. These attacks begin with an email and utilize social engineering techniques to trick individuals into clicking on a malicious link. Researchers have proposed various techniques, including machine learning models, to improve the detection of phishing URLs. This article presents the findings of an experimental study that demonstrates the effectiveness of data balancing, hyperparameter optimization, and feature selection in enhancing the accuracy of machine learning algorithms. The results show significant improvements in accuracy and outperform existing research works.
Article
Computer Science, Information Systems
Raman Singh, Sean Sturley, Hitesh Tewari
Summary: The utilization of Internet of Things (IoT) devices in smart city and industrial applications is increasing rapidly. A group authentication framework is proposed to establish trust between IoT devices owned by different entities. Blockchain technology is used to support the secure exchange of information between trusted authorities.
Article
Chemistry, Analytical
M. Priyadharshini, A. Faritha Banu, Bhisham Sharma, Subrata Chowdhury, Khaled Rabie, Thokozani Shongwe
Summary: In recent years, machine learning and computer vision have grown in their use of multi-label categorization. Research has used SMOTE for data balance, but it can lead to more class overlap and noise. To address this, a technique called Adaptive Synthetic Data-Based Multi-label Classification (ASDMLC) is introduced. It utilizes ADASYN sampling for learning from unbalanced datasets and employs Velocity-Equalized Particle Swarm Optimization (VPSO) for feature selection to improve multi-label classification accuracy. The proposed approach combines Adaptive Neuro-Fuzzy Inference System (ANFIS), Probabilistic Neural Network (PNN), and Clustering-Based Decision tree methods for label dependencies. The suggested model achieves a multi-label classification accuracy of 90.88%, surpassing previous techniques.
Editorial Material
Chemistry, Analytical
Bhisham Sharma, Deepika Koundal, Rabie A. Ramadan, Juan M. Corchado
Article
Materials Science, Multidisciplinary
X. Anitha Mary, Bhisham Sharma, I. Johnson, J. Chalmers, C. Karthik, Subrata Chowdhury
Summary: Groundwater is widely used in various sectors and is considered an essential natural resource. This study aims to develop a machine learning model to forecast groundwater quality in Tamil Nadu, India, using a comprehensive dataset of groundwater attributes. Various ML regression algorithms were compared to predict water quality index (WQI), and the results were evaluated using performance metrics. The ensemble model (EM) showed the lowest root mean square error (RMSE) of 2.4x10(-6). The predicted WQI values were also used to classify different districts in Tamil Nadu.
INTERNATIONAL JOURNAL OF COMPUTATIONAL MATERIALS SCIENCE AND ENGINEERING
(2023)
Article
Environmental Sciences
Kashif Naseer Qureshi, Ayesha Khan, Syed Umair Ullah Jamil, Bhisham Sharma, Gwanggil Jeon
Summary: Solid waste management is an urgent concern that requires attention from citizens and government stakeholders. This paper analyzes the challenges faced in waste management and proposes a technological solution to address these issues. The solution utilizes sensor-based technologies, cellular networks, and social media to monitor waste in urban areas and ensure efficient collection. The experiment results indicate the effectiveness of the proposed solution.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Automation & Control Systems
Deepika Saravagi, Shweta Agrawal, Manisha Saravagi, Sanjiv K. Jain, Bhisham Sharma, Abolfazl Mehbodniya, Subrata Chowdhury, Julian L. Webber
Summary: This study proposes an integrated model, S-VCNet, which combines VGG16 and CapsNet for the identification of spondylolisthesis. The model uses VGG16 as a feature extractor and then feeds the output to CapsNet for disease identification. Experimental results demonstrate that the developed model achieves an accuracy of 98% in lumbar spondylolisthesis diagnosis.
INTELLIGENT AUTOMATION AND SOFT COMPUTING
(2023)
Article
Environmental Sciences
Flora Gnanadhas, Surendarnath Sundaramoorthy, Sowndharya Natarajan, Mary Stephy Gnanamanickam, Kassian T. T. Amesho, Bhisham Sharma
Summary: This study investigated the extraction and dyeing properties of natural fabric dyes derived from brown seaweeds. The results showed that aqueous and ethanol dye extracts exhibited superior fastness properties compared to acetone and methanol extracts. Additionally, this study explored the bioactive potential of natural fabric dyes derived from brown seaweeds, providing a sustainable alternative to synthetic dyes.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Computer Science, Information Systems
Luong Thi Theu, Tran Quang-Huy, Tran Duc-Tan, Bhisham Sharma, Subrata Chowdhury, Karthik Chandran, Saravanakumar Gurusamy
Summary: This paper proposes two enhanced schemes to tackle the challenge of the undetermined inverse problem in pharmacokinetics, in order to find multiple possible point sets for drug kinetics in the patient's body. The first scheme applies Tikhonov regularization to deal with the overdetermined system, while the second scheme employs perturbation-level tuning for numerical stabilization. Numerical simulations show that these schemes can reduce iterations and computation time, and enable PoC to move steadily towards the solutions manifold.