Article
Computer Science, Information Systems
Ahmed Omar, Tarek M. Mahmoud, Tarek Abd-El-Hafeez, Ahmed Mahfouz
Summary: Online Social Networks (OSNs) are popular for communication and sharing personal information, but they also face challenges such as inappropriate content. Research on Arabic language in this field is limited.
INFORMATION SYSTEMS
(2021)
Article
Computer Science, Information Systems
Francisco de Arriba-Perez, Silvia Garcia-Mendez, Francisco J. Gonzalez-Castano, Jaime Gonzalez-Gonzalez
Summary: Artificial Intelligence techniques, specifically Machine Learning, have not been fully utilized in the legal domain due to the lack of explanatory capabilities. This research proposes a hybrid system that combines ML for multi-label classification of legal judgements with visual and natural language explanations. By incorporating Natural Language Processing techniques and deep legal reasoning, the system is able to identify entities involved in the cases. The solution achieves over 85% micro precision and shows potential to alleviate labor-intensive legal classification tasks.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Ye Jiang, Yimin Wang
Summary: This study proposes T-HMAN, a Topic-aware Hierarchical Multiple Attention Network, for text classification. The model combines a multi-head self-attention mechanism with convolutional filters to capture long-range semantic dependency, and integrates topic distributions generated by LDA with sentence-level and document-level inputs in a hierarchical architecture. The proposed model outperforms current state-of-the-art hierarchical models on five publicly accessible datasets.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Review
Computer Science, Artificial Intelligence
Ahlam Wahdan, Mostafa Al-Emran, Khaled Shaalan
Summary: This article provides an overview of the current status of Arabic text classification, including research methods, topic areas, and application tasks. The study finds that there is a lack of thorough evaluation of Arabic text classification. The article proposes directions for further research, such as addressing the issue of unbalanced datasets and improving the preprocessing phase.
Review
Chemistry, Multidisciplinary
Ali Saleh Alammary
Summary: BERT has gained attention for its unique features in natural language processing, with the introduction of models supporting different languages like Arabic. However, the current state of applying BERT to Arabic text classification is limited, prompting the need for further research and improvements.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Jiufeng Zhao, Rui Song, Chitao Yue, Zhenxin Wang, Hao Xu
Summary: With the development of e-government in China, multiple local governments are creating online platforms that require automatic policy classification. However, current methods for policy classification rely on supervised models and annotated policies, which are expensive and difficult to obtain. To address this, a large-scale framework called Weak-PMLC is proposed, which utilizes weak supervision and label names to generate pseudo-labeled policies for training. This method achieves high performance and outperforms existing weakly supervised methods.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Multidisciplinary Sciences
Hanen Himdi, George Weir, Fatmah Assiri, Hassanin Al-Barhamtoshy
Summary: This paper addresses the issue of detecting fake news in the Arabic language. It introduces a supervised machine learning model for classifying Arabic news articles based on their context's credibility and presents the first dataset of Arabic fake news articles generated through crowdsourcing. The findings show that the performance of this model outperforms humans in the same task.
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Information Systems
Badriyya B. Al-Onazi, Hassan Alshamrani, Fatimah Okleh Aldaajeh, Amira Sayed A. Aziz, Mohammed Rizwanullah
Summary: This study develops a deep learning-based technique for sentiment classification of Arabic tweets, effectively recognizing and categorizing emotions in Arabic text data.
Article
Computer Science, Artificial Intelligence
Boyan Wang, Xuegang Hu, Peipei Li, Philip S. Yu
Summary: The paper proposes a unified framework, Hierarchical Cognitive Structure Learning Model (HCSM), for handling hierarchical multi-label text classification (HMLTC) tasks. This model comprehensively utilizes partial new knowledge and global hierarchical label structure, demonstrating superior performance in experimental results on four benchmark datasets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Yinglong Ma, Xiaofeng Liu, Lijiao Zhao, Yue Liang, Peng Zhang, Beihong Jin
Summary: This paper introduces a hierarchical multi-label text classification method based on hybrid embedding, combining graph embedding and word embedding; using a level-by-level HMTC approach and conducting extensive experiments on five large-scale real-world datasets, the results show that the method is competitive in classification accuracy.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Theory & Methods
Jose Antonio Garcia-Diaz, Mar Canovas-Garcia, Ricardo Colomo-Palacios, Rafael Valencia-Garcia
Summary: Online social networks empower individuals to harass others anonymously, with women being frequently targeted. Despite efforts to combat misogyny, it remains challenging to identify due to its subtle nature and cultural variations.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Abdullah Y. Muaad, Shaina Raza, Usman Naseem, Hanumanthappa J. Jayappa Davanagere
Summary: After becoming the sixth official language of the United Nations, the Arabic language plays a crucial role in the world. However, due to the complexity of morphology and diversity of Arabic dialects, existing algorithms are not effective for processing Arabic text. This study surveys research on Arabic Text Detection (ATD) from 2017 to 2023, identifying several areas of interest that still need to be addressed. Deep-based methods, although slower, offer better results by comprehending both the context and semantics of the language. Hybrid models show promise as a way forward for future research.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Meng Xiao, Ziyue Qiao, Yanjie Fu, Hao Dong, Yi Du, Pengyang Wang, Hui Xiong, Yuanchun Zhou
Summary: The increasing interdisciplinary nature of research proposals poses a challenge in assigning appropriate reviewers. This study aims to develop a fair and precise proposal reviewer assignment system by leveraging AI. A deep hierarchical interdisciplinary research proposal classification network (HIRPCN) is proposed, which extracts textual semantic information, learns interdisciplinary knowledge, and detects interdisciplinary topic paths for each proposal. Extensive experiments and expert evaluations demonstrate the effectiveness of the proposed model.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Aladdin Masri, Muhannad Al-Jabi
Summary: Over the last few decades, there has been a significant increase in reliance on email communication, especially in business. As a result, researchers have proposed algorithms to classify and redistribute the large number of emails that companies receive daily. However, there is limited literature on Arabic text classification, despite the growing concern over Arabic emails in official correspondence. This study addresses this gap by applying natural language processing to classify Arabic business emails. Using a dataset of 63,257 emails, the proposed models, based on machine learning techniques and a lexicon of words, achieved promising results with an accuracy of about 92% and a loss of less than 8%. This work demonstrates the correctness and robustness of the models.
PEERJ COMPUTER SCIENCE
(2023)
Article
Computer Science, Information Systems
Fatima-zahra El-Alami, Said Ouatik El Alaoui, Noureddine En Nahnahi
Summary: This paper investigates the potential of pre-trained Arabic BERT model for learning contextual sentence representations and its application in Arabic text multi-class categorization. Experimental results show that fine-tuned AraBERT model achieves state-of-the-art performance.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Telecommunications
Nusaybah Alghanmi, Reem Alotaibi, Seyed M. Buhari
Summary: This paper reviews the relevant literature on anomaly detection techniques using various machine learning approaches in the IoT, analyzes the issues with different anomaly detection datasets, and lists future research directions in this field.
WIRELESS PERSONAL COMMUNICATIONS
(2022)
Article
Mathematics, Interdisciplinary Applications
Ximei Li, Reem Alotaibi
Summary: The purpose of this paper is to construct different risk measurement models by combining the financing risk of enterprises with the psychological factors of the consumer market, in order to evaluate the pricing and financing risks of technology-based SMEs. The findings show that the nonlinear expectation and stochastic differential equation can reflect changes in enterprise value and the impact of investor psychology on financing effectiveness. By applying the nonparametric estimation method, the accuracy of the model prediction can be improved.
FRACTALS-COMPLEX GEOMETRY PATTERNS AND SCALING IN NATURE AND SOCIETY
(2022)
Article
Computer Science, Artificial Intelligence
Madini O. Alassafi, Mutasem Jarrah, Reem Alotaibi
Summary: The study developed a prediction model for the spread of COVID-19 in Malaysia, Morocco, and Saudi Arabia using public datasets from the European Centre for Disease Prevention and Control. Deep learning models were utilized with a focus on LSTM networks. The study also compared the number of cases and deaths in the three countries.
Article
Environmental Sciences
Kamil Faisal, Sultanah Alshammari, Reem Alotaibi, Areej Alhothali, Omaimah Bamasag, Nusaybah Alghanmi, Manal Bin Yamin
Summary: The spatial distribution of vaccine centers is crucial for effective epidemic responses, and GIS analysis can be used to enhance coverage and efficiency. It is recommended to consider areas with broader coverage when allocating vaccine centers and to increase the number of centers to ensure fairness and equity in vaccine distribution.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Review
Public, Environmental & Occupational Health
Nusaybah Alghanmi, Reem Alotaibi, Sultanah Alshammari, Areej Alhothali, Omaimah Bamasag, Kamil Faisal
Summary: This study presents a survey of the point of dispensing (PODs) location-allocation problem during public health emergencies. The survey analyzes existing models based on full and partial demand points allocation and compares them based on their features, strengths, and limitations. The study also discusses the challenges and future research directions for PODs location-allocation models. The results highlight the need for developing techniques to meet the demands of specific groups and to consider country-specific variations in population size and density.
FRONTIERS IN PUBLIC HEALTH
(2022)
Article
Computer Science, Artificial Intelligence
Sahar Aldhaheri, Reem Alotaibi, Bandar Alzahrani, Anas Hadi, Arif Mahmood, Areej Alhothali, Ahmed Barnawi
Summary: This paper proposes a multi-task attention based crowd counting network (MACC Net) to address the challenges in crowd density estimation. The network improves counting accuracy through density level classification, density map estimation, and segmentation guided attention. Experimental results on multiple datasets demonstrate that the MACC Net achieves state of the art performance in crowd counting.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Areej Alhothali, Amal Balabid, Reem Alharthi, Bander Alzahrani, Reem Alotaibi, Ahmed Barnawi
Summary: Recognizing and localizing anomalous events in crowd scenes is a challenging problem. This research aims to detect and locate anomalies in dense crowd scenes, proposing a method that combines deep learning with support vector machines.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Madini O. Alassafi, Muhammad Sohail Ibrahim, Imran Naseem, Rayed AlGhamdi, Reem Alotaibi, Faris A. Kateb, Hadi Mohsen Oqaibi, Abdulrahman A. Alshdadi, Syed Adnan Yusuf
Summary: The vulnerability of conventional face recognition systems to face presentation or face spoofing attacks has attracted attention. Deep learning-based face presentation attack detection (PAD) methods have gained popularity. This research proposes a supervised contrastive learning approach to tackle the face anti-spoofing problem.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Tarik Alafif, Anas Hadi, Manal Allahyani, Bander Alzahrani, Areej Alhothali, Reem Alotaibi, Ahmed Barnawi
Summary: Individual abnormal behaviors vary depending on crowd sizes, contexts, and scenes. Challenges occur in large-scale crowds when detecting, tracking, and recognizing individuals with abnormalities. This paper introduces a large-scale crowd abnormal behavior dataset and proposes a method using hybrid CNNs and RFs to detect and recognize abnormal behaviors.
Article
Computer Science, Artificial Intelligence
Miada Almasri, Norah Al-Malki, Reem Alotaibi
Summary: This research aims to enhance the capability of a deep learning model, AraBERT v02, for aspect category detection in the Arabic language. The study utilizes a semi-supervised self-training approach called the noisy student framework. Findings show that the ensembled teacher-student model outperforms baselines and other deep learning models in predicting aspect categories.
PEERJ COMPUTER SCIENCE
(2023)
Article
Engineering, Multidisciplinary
Reem Alotaibi, Felwa Abukhodair
Summary: Radiation dose tracking is important due to the popularity of CT scans. However, existing software programs for tracking doses have limitations and do not provide accurate answers. This paper proposes a visual analytic approach using Tableau software to track radiation dose data from CT scans, which had a 100% success rate in real-life scenarios and improved the tracking process.
Article
Computer Science, Information Systems
Madini O. Alassafi, Muhammad Sohail Ibrahim, Imran Naseem, Rayed AlGhamdi, Reem Alotaibi, Faris A. Kateb, Hadi Mohsen Oqaibi, Abdulrahman A. Alshdadi, Syed Adnan Yusuf
Summary: Face presentation attack detection (PAD) is a crucial step in modern face recognition systems to expose imposters and unauthorized persons. This research proposes a novel face PAD solution using interpolation-based image diffusion and transfer learning of a MobileNet convolutional neural network. The experimental results show that the proposed method outperforms most state-of-the-art methods in terms of performance.
Article
Mathematics, Applied
Fanxiu Gao, Reem Alotaibi, Mohammed Yousuf Abo Keir
Summary: This article introduces an improved sales percentage method and uses SPSS for regression analysis to predict future sales revenue, calculate future net cash flow, and company value based on the predicted data.
APPLIED MATHEMATICS AND NONLINEAR SCIENCES
(2022)
Article
Mathematics, Applied
Xin Guan, Peng Yao, Reem Alotaibi, Mohammed Yousuf Abo Keir
Summary: This paper explains the numerical instability in engineering structure topology optimization using the finite element method. The Gaussian function filtering method is introduced to reduce the global impact of local extremum in the optimization process, and successfully applied in the topology optimization of building structures in hydraulic engineering.
APPLIED MATHEMATICS AND NONLINEAR SCIENCES
(2022)
Article
Computer Science, Information Systems
Bander Alzahrani, Ahmed Barnawi, Azeem Irshad, Areej Alhothali, Reem Alotaibi, Muhammad Shafiq
Summary: Unmanned aerial vehicles (UAVs) have gained significant attention in civil and commercial applications, especially in the field of crowd monitoring. However, ensuring the security and privacy of communication between drones and controlling entities remains a critical challenge. This paper proposes an enhanced authenticated key agreement (AKA) solution for secure communication, and its effectiveness is demonstrated through simulation and verification.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)