Article
Computer Science, Artificial Intelligence
Jameel Almalki
Summary: This study focuses on sentiment analysis of Arabic tweets in the distance learning domain in Saudi Arabia, and proposes a model based on Apache Spark. By retrieving tweets through the Twitter API and preprocessing them, sentiment prediction is performed using a Logistic Regression model. The proposed model demonstrates superior performance in sentiment analysis of Arabic tweets.
PEERJ COMPUTER SCIENCE
(2022)
Review
Multidisciplinary Sciences
Andrea Manconi, Matteo Gnocchi, Luciano Milanesi, Osvaldo Marullo, Giuliano Armano
Summary: Advances in high-throughput and digital technologies have led to the adoption of big data in life sciences. However, utilizing big data presents technical and infrastructural challenges. Apache Spark, a powerful HPC engine, offers a solution for large-scale data processing and machine learning applications.
Article
Computer Science, Artificial Intelligence
Hamidreza Kadkhodaei, Amir Masoud Eftekhari Moghadam, Mehdi Dehghan
Summary: This paper presents a distributed heterogeneous ensemble classifier for big data, which utilizes multiple classifiers to achieve more accurate data classification. Experimental results indicate the superiority of the proposed method in terms of classification accuracy, performance, and scalability compared to existing ensemble algorithms.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Chemistry, Multidisciplinary
Mladen Amovic, Miro Govedarica, Aleksandra Radulovic, Ivana Jankovic
Summary: Smart cities leverage digital technologies to enhance geospatial data exchange, boost data utilization, and establish new services for sustainable development. The GAMINESS management system, based on big data modeling principles, efficiently stores and manages vast amounts of structured, semi-structured, and unstructured data in real time, improving system performance through the five V principles.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Interdisciplinary Applications
Muralidhar Kurni, Mujeeb S. Md, Bharath Bhushan Yannam, T. Arun Singh
Summary: The paper introduces an efficient intrusion detection method called MRPO-based Deep maxout network, which combines various algorithms for data processing and feature selection to improve detection accuracy. The proposed method demonstrates superior performance compared to traditional approaches in practical applications.
ADVANCES IN ENGINEERING SOFTWARE
(2022)
Article
Computer Science, Artificial Intelligence
Chaoyu Gong, Zhi-gang Su, Pei-hong Wang, Yang You
Summary: The study introduces a distributed evidential clustering algorithm that can analyze time series data on a large scale without losing information, improving the accuracy and interpretability of clustering results.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Hardware & Architecture
Tinku Singh, Riya Khanna, Satakshi, Manish Kumar
Summary: This study proposes an efficient method for dealing with the multi-class classification of imbalanced large datasets in Apache Spark. By utilizing an improved version of the synthetic minority oversampling technique (SMOTE), it generates sufficient synthetic data points for the minority classes. Experimental results demonstrate the effectiveness of this method in classifying unknown data samples from large datasets.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Information Systems
Reem Hamed Aljuhani, Areej Alshutayri, Shahd Alahdal
Summary: This study focuses on emotion recognition based on Arabic Saudi dialect spoken data, utilizing various spectral features and applying different classification methods. Experimental results showed that Support Vector Machine (SVM) achieved the best accuracy in Arabic speech emotion recognition, demonstrating improvement in this classification method.
Article
Chemistry, Analytical
Malak Aljabri, Sara Mhd. Bachar Chrouf, Norah A. Alzahrani, Leena Alghamdi, Reem Alfehaid, Reem Alqarawi, Jawaher Alhuthayfi, Nouf Alduhailan
Summary: The Ministry of Education in Saudi Arabia implemented distance learning to cope with the COVID-19 pandemic, and a study was conducted to analyze people's attitudes towards this learning method on social media. The research found generally positive opinions towards distance learning for kindergarten, intermediate, and high school stages, while negative opinions were observed for the university stage.
Article
Computer Science, Information Systems
Abderrahmane Ed-daoudy, Khalil Maalmi, Aziza El Ouaazizi
Summary: This study proposes a scalable and real-time system for disease prediction by integrating Twitter, Apache Kafka, Apache Spark, and Apache Cassandra. The heart disease dataset was used for experiments, and the Relief algorithm was applied for features selection, resulting in a classification accuracy of 92.05%. The performance comparison of different machine learning algorithms implemented in Spark MLlib showed that Spark outperformed in scalability and computing times.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Theory & Methods
Wadee Alhalabi, Jari Jussila, Kamal Jambi, Anna Visvizi, Hafsa Qureshi, Miltiadis Lytras, Areej Malibari, Raniah Samir Adham
Summary: The use of social media has increased dramatically in recent years, posing critical requirements for behavior supervision and fraud protection like detecting terrorist behavior. A key priority project called ALT-TERROS, an artificial intelligence-enabled system for detecting terrorist behavior, has been developed with requirements including data integration, advanced smart analysis capacity, and decision-making capability. The project offers a sophisticated integrated approach for managing distributed data on social media and detecting patterns of terrorist behavior using advanced social mining methods, which can be visualized and used for decision-making.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Chemistry, Multidisciplinary
Nikitha Johnsirani Venkatesan, Dong Ryeol Shin, Choon Sung Nam
Summary: This study focuses on improving the accuracy of lung nodule detection by utilizing a convolutional neural network model and removing noise. Experimental results show superior performance on the LUNA16 dataset, and comparison on various platforms indicates that Apache Spark is suitable for parallel training computation with high accuracy.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Ahmed Ismail Ebada, Ibrahim Elhenawy, Chang-Won Jeong, Yunyoung Nam, Hazem Elbakry, Samir Abdelrazek
Summary: This study aims to develop a real-time healthcare data prediction system using big data processing, enabling the prediction of health issues and sending alerts and recommendations to users and healthcare providers. The proposed system also aims to provide an effective recommendation system by utilizing streaming medical data, historical data, and a knowledge database to generate real-time suggestions and alerts through machine learning algorithms.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Computer Science, Information Systems
Yelleti Vivek, Vadlamani Ravi, P. Radha Krishna
Summary: In this study, two parallel and distributed hybrid evolutionary algorithm wrappers based on Apache Spark were proposed to address the scalability and performance issues of existing feature subset selection algorithms when applied to big datasets. The PB-TADE algorithm demonstrated significant improvement and achieved a speedup of 2.2-2.9 compared to other algorithms.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Nadiah A. Baghdadi, Amer Malki, Hossam Magdy Balaha, Yousry AbdulAzeem, Mahmoud Badawy, Mostafa Elhosseini
Summary: Many people suffer from mental illnesses like major depressive disorder (MDD), and Twitter can be used to track depression. However, depressive detection methods for Arabic tweets have not been applied. This study proposes a classification framework and introduces an Arabic tweet preprocessing algorithm. Experiments show that Arabic BERT models perform better than USE models.
PEERJ COMPUTER SCIENCE
(2022)
Article
Computer Science, Information Systems
Abubakar Sadiq Mohammed, Eirini Anthi, Omer Rana, Neetesh Saxena, Pete Burnap
Summary: Industrial Cyber-Physical Systems (ICPS) rely on Supervisory Control and Data Acquisition (SCADA) for process monitoring and control. However, communication through insecure protocols such as Modbus, DNP3, and OPC Data Access makes these SCADA systems vulnerable to various attacks, including denial of service (DoS) attacks. This paper introduces a novel Field Flooding attack that exploits the packet memory structure of the Modbus protocol to perform a DoS attack on Programmable Logic Controllers (PLCs). The proposed mechanism, utilizing supervised machine learning with the XGBoost algorithm, achieves 99% accuracy in detecting this attack.
COMPUTERS & SECURITY
(2023)
Article
Computer Science, Software Engineering
Adnan Umer, Adnan Noor Mian, Omer Rana
Summary: This study analyzes machine failure and changes in machine lifecycle using Google cluster trace data set. A Markov chain-based model is proposed to predict machine states, and it is validated with actual data, showing a small error.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2023)
Article
Computer Science, Theory & Methods
Lamya Alkhariji, Suparna De, Omer Rana, Charith Perera
Summary: This research aims to develop a personal assistant that can provide answers to software engineers' questions about privacy protection during the design phase of IoT system development. The developed PARROT ontology, based on semantic web technologies and representative IoT use cases, can answer up to 58% of privacy-related questions.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2023)
Article
Computer Science, Theory & Methods
Zichuan Xu, Dapeng Zhao, Weifa Liang, Omer F. Rana, Pan Zhou, Mingchu Li, Wenzheng Xu, Hao Li, Qiufen Xia
Summary: This article investigates the cost minimization problem of implementing federated learning requests in a mobile edge computing network and proposes an optimization framework and algorithms. The performance of the proposed algorithms is evaluated through simulations and experiments, showing that they outperform benchmark algorithms in reducing implementation cost.
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
(2023)
Article
Mathematics
Iyad Katib, Mahmoud Ragab
Summary: The centralized storage structure in IoT applications poses security, privacy, and single point of failure issues, which can be addressed by using blockchain technology. However, DDoS attacks have revealed limitations in blockchain-assisted IoT networks. This study proposes a hybrid Harris Hawks with sine cosine and deep learning-based intrusion detection system (H3SC-DLIDS) to recognize DDoS attacks in a blockchain-supported IoT environment.
Article
Computer Science, Information Systems
Muhammed Golec, Sukhpal Singh Gill, Mustafa Golec, Minxian Xu, Soumya K. Ghosh, Salil S. Kanhere, Omer Rana, Steve Uhlig
Summary: This paper presents a framework called BlockFaaS that integrates serverless platform and blockchain architecture to support dynamic scalability, security, and privacy in AI-based healthcare applications. The results show that BlockFaaS outperforms other frameworks in terms of performance.
JOURNAL OF GRID COMPUTING
(2023)
Article
Computer Science, Hardware & Architecture
Sadoon Azizi, Pedram Farzin, Mohammad Shojafar, Omer Rana
Summary: Provisioning services for IoT devices faces challenges such as device heterogeneity, diverse QoS requirements, and availability of Cloud and Fog resources. FLEX is proposed as a platform for service placement in multi-Fog and multi-Cloud environments, which broadcasts service requirements and selects suitable providers based on factors like resources, location, QoS, and cost. The proposed algorithm, MCD1, shows significant performance improvements in reducing delay and cost compared to baseline methods and genetic algorithms.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Automation & Control Systems
Gang Qin, An Lin, Jun Cheng, Mengjie Hu, Iyad Katib
Summary: This study investigates the issue of event-triggered fault detection filtering for memristive neural networks with dynamic quantization in the discrete-time domain. A novel event-triggered protocol is proposed based on dynamic quantization parameter, fault occurrence probability, and network bandwidth utilization rate, and an asynchronous filter framework is developed to ensure the stability of the system.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2023)
Article
Automation & Control Systems
Wei Kang, Gang Qin, Jun Cheng, Huaicheng Yan, Iyad Katib, Jinde Cao
Summary: This paper proposes a security control method for a discrete-time switched power system using a probabilistic event-triggered protocol, which effectively optimizes network resource utilization and improves system security and stability under multi-strategy deception attacks.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Junhui Wu, Gang Qin, Jun Cheng, Jinde Cao, Huaicheng Yan, Iyad Katib
Summary: This paper proposes an innovative approach to mitigate the effects of deception attacks in Markov jumping systems by developing an adaptive neural network control strategy. The approach effectively approximates the unbounded false signals injected by deception attacks and establishes a connection between the joint Markov chain and controller.
Article
Computer Science, Information Systems
Iyad Katib, Fatmah Y. Assiri, Turki Althaqafi, Zenah Mahmoud Alkubaisy, Diaa Hamed, Mahmoud Ragab, Heung-Il Suk
Summary: Smart Fintech, empowered by data science and artificial intelligence, drives automated, intelligent, personalized financial and economic businesses, playing a crucial role in today's technology-driven society and economies.
Review
Computer Science, Information Systems
Sundas Iftikhar, Sukhpal Singh Gill, Chenghao Song, Minxian Xu, Mohammad Sadegh Aslanpour, Adel N. Toosi, Junhui Du, Huaming Wu, Shreya Ghosh, Deepraj Chowdhury, Muhammed Golec, Mohit Kumar, Ahmed M. Abdelmoniem, Felix Cuadrado, Blesson Varghese, Omer Rana, Schahram Dustdar, Steve Uhlig
Summary: This paper conducts a systematic literature review to analyze the applicability and challenges of AI and ML algorithms in resource management for fog/edge computing environments. Various machine learning, deep learning, and reinforcement learning techniques for edge AI management are discussed. A taxonomy of AI/ML-based resource management techniques for fog/edge computing is proposed, and existing techniques are compared based on this taxonomy. Open challenges and promising future research directions in AI/ML-based fog/edge computing are also identified and discussed.
INTERNET OF THINGS
(2023)
Article
Computer Science, Information Systems
Nada Alhirabi, Stephanie Beaumont, Jose Tomas Llanos, Dulani Meedeniya, Omer Rana, Charith Perera
Summary: IoT applications often involve the collection and analysis of sensitive personal data, which requires higher levels of protection. However, software developers commonly neglect privacy practices. To address this issue, a design tool called PARROT is introduced, which helps developers design privacy-aware IoT applications and offers real-time feedback on potential privacy violations.
PROCEEDINGS OF THE ACM ON INTERACTIVE MOBILE WEARABLE AND UBIQUITOUS TECHNOLOGIES-IMWUT
(2023)
Article
Computer Science, Information Systems
Victor Medel, Unai Arronategui, Omer Rana, Jose Angel Banares, Rafael Tolosana-Calasanz
Summary: Performance interference can occur in cloud systems when different services are executed on the same physical infrastructure, resulting in degraded performance. This study proposes a CFA-based model to estimate performance interference across containers caused by the use of CPU, memory, and IO. The model provides resource characterization through human comprehensible indices expressed as time series, enabling analysis of interference throughout the service execution lifetime. Experiments using real services executed in Docker containers show that the model accurately predicts overall execution time for different service combinations. This approach allows service designers to identify phases likely to cause greater interference and ensure only complementary services are hosted together, enabling more intelligent resource management and scheduling.
Article
Computer Science, Information Systems
Musfirah Ihtesham, Shahzaib Tahir, Hasan Tahir, Anum Hasan, Aiman Sultan, Saqib Saeed, Omer Rana
Summary: Serverless computing has experienced significant growth due to its adaptability and deployment agility, leading to an increased need for security measures such as searchable encryption. This paper introduces a novel privacy-preserving multiple keyword searchable encryption scheme within a serverless cloud environment, utilizing probabilistic encryption and fully homomorphic encryption to enable secure and efficient searches on encrypted data. Through rigorous testing on a real-world dataset, the proposed scheme demonstrates robust security and usability in preserving the privacy of search queries and access patterns.