Article
Computer Science, Artificial Intelligence
Ahmed Abdelfattah Saleh, Weigang Li
Summary: This research proposes a language agnostic automatic text summarization model called TxLASM, which relies on the characteristics of major text elements to perform extractive summarization without the need for language dependent preprocessing tools. The model encodes the shapes of text elements, extracts their intrinsic features, and generates language, domain, and context independent summaries.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Artificial Intelligence
Mengyun Cao, Hai Zhuge
Summary: This research proposes a general framework for automatically evaluating the informativeness, conciseness, and coherence of summaries. The framework utilizes a semantic link network to represent and analyze texts, achieving comparable or higher correlations with human judgments compared to popular evaluation models.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Thierry S. Barros, Carlos Eduardo S. Pires, Dimas Cassimiro Nascimento
Summary: A document called notitia criminis is used by the Brazilian Federal Police as the starting point of criminal investigations, containing all relevant information about the supposed crime. To extract essential information from these documents, which can be mentally exhausting due to their size and complexity, the BERT model is proposed as a solution. Experimental results using different variants of the ROUGE metric demonstrate the effectiveness and efficiency of the BERT-based approaches in extractive text summarization.
KNOWLEDGE AND INFORMATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Ozan Ozyegen, Devika Kabe, Mucahit Cevik
Summary: This paper demonstrates how different text highlighting techniques can alleviate the workload of medical professionals and improve the quality of online medical services. The numerical study shows that the neural network approach is successful in highlighting medically-relevant terms.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2022)
Review
Computer Science, Information Systems
Deepali Jain, Malaya Dutta Borah, Anupam Biswas
Summary: The paper emphasizes the importance of extensive research in legal text processing, focusing on various text summarization techniques and tools for legal document summarization. It presents case studies on automatic summarization of legal documents from different countries, aiming to provide researchers with a starting point for in-depth exploration in this field.
COMPUTER SCIENCE REVIEW
(2021)
Review
Computer Science, Artificial Intelligence
Lingfeng Lu, Yang Liu, Weiqiang Xu, Huakang Li, Guozi Sun
Summary: Automatic summarization is a promising research area that has gained increasing attention. It has been applied in various real-world scenarios with positive results. However, conventional evaluation metrics are not keeping up with the evolving summarization task formats and indicators. Recent research has shown that automatic summarization requires not only readability and fluency, but also informativeness and consistency. Diversified application scenarios pose new challenges for generative language models and evaluation metrics. In this review, we analyze and focus on the differences between the task format and the evaluation metrics.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Artificial Intelligence
Bing Ma, Hai Zhuge
Summary: This paper proposes a method to represent texts using common words and measure the similarity of text classes. It also introduces a bottom-up text clustering approach to construct class trees. Experimental results show that this method outperforms other algorithms in terms of classification accuracy and class tree structure. Additionally, a document summarization approach based on this method achieves good performance.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Artificial Intelligence
H. Abo-Bakr, S. A. Mohamed
Summary: Due to the vast amount of textual information, automatic text summarization (ATS) systems are necessary for extracting important information or generating summaries. This work proposes an extractive ATS system that preprocesses the text and formulates the summarization as a multi-objective optimization problem. An evolutionary sparse multi-objective algorithm is developed to solve the problem and produce a set of non-dominated summaries. The system has been evaluated using DUC datasets and compared to existing literature using ROUGE metrics.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Computer Science, Information Systems
Deepa Anand, Rupali Wagh
Summary: The availability of legal judgment documents in digital form opens up opportunities for information extraction and application. This paper proposes generic techniques using neural network for summarizing Indian legal judgment documents. The proposed approaches do not rely on handcrafted features or domain-specific knowledge, making them suitable for other domains as well.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Mohammad Bani-Almarjeh, Mohamad-Bassam Kurdy
Summary: Recently, research has shown that Transformer model architecture and pre-trained Transformer-based language models perform well in natural language understanding and text generation. However, there is limited research in using these models for text generation in Arabic. This study aims to evaluate and compare different model architectures and pre-trained language models for Arabic abstractive summarization. Results show that Transformer-based models significantly outperform traditional RNN-based models and using less data. AraT5, a encoder-decoder pre-trained Transformer, is found to be more suitable for summarizing Arabic text compared to the AraBERT-initialized BERT2BERT model. Additionally, both AraT5 and AraGPT2 perform better than AraBERT in summarizing out-of-domain text.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Artificial Intelligence
Xin Shen, Wai Lam, Shumin Ma, Huadong Wang
Summary: Recently, the attention of researchers has been drawn to neural abstractive text summarization (NATS) models based on sequence-to-sequence architecture. However, these models face limitations in summarizing long documents. In this paper, a novel NATS framework called UOTSumm is proposed, which learns text alignment directly from summarization data without using biased tools like ROUGE. Experimental results show that UOTSumm outperforms other models that rely on ROUGE for text alignment on three publicly available LDS benchmarks.
NATURAL LANGUAGE ENGINEERING
(2023)
Review
Chemistry, Multidisciplinary
Nikolaos Giarelis, Charalampos Mastrokostas, Nikos Karacapilidis
Summary: Text summarization is the automatic creation of a concise and fluent summary capturing the main ideas and topics from one or multiple documents. Extractive approaches rank and combine important sentences, while abstractive approaches generate summaries with new phrases and sentences. However, both approaches lack the contextual representation needed for fluent summaries. This survey provides a comprehensive evaluation framework, including a survey of state-of-the-art approaches, a comparative evaluation using popular evaluation scores, insights on various aspects of summarization, and the release of datasets and code for reproducibility.
APPLIED SCIENCES-BASEL
(2023)
Article
Biochemical Research Methods
Meng Zhang, Madhuri Sankaranarayanapillai, Jingcheng Du, Yang Xiang, Frank J. Manion, Marcelline R. Harris, Cooper Stansbury, Huy Anh Pham, Cui Tao
Summary: This study demonstrates the feasibility of using machine learning tools to classify permission-related sentences in informed consent documents.
BMC BIOINFORMATICS
(2023)
Review
Computer Science, Artificial Intelligence
Wafaa S. El-Kassas, Cherif R. Salama, Ahmed A. Rafea, Hoda K. Mohamed
Summary: Automatic Text Summarization (ATS) is crucial due to the exponential growth of textual content on the Internet, but generated summaries still fall short of human-generated ones. Researchers need to focus more on abstractive and hybrid approaches.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Yogesh Kumar, Komalpreet Kaur, Sukhpreet Kaur
Summary: In today's world, there is an abundance of information available online and offline, making it challenging to manually extract useful information from numerous articles on a single topic. Automatic text summarization systems have been developed to address this issue, extracting and compressing essential information from large documents into concise summaries. This survey paper provides a comprehensive overview of research on automatic text summarization in various languages, offering valuable insights and technical knowledge for beginner scientists in this field.
ARTIFICIAL INTELLIGENCE REVIEW
(2021)
Article
Computer Science, Artificial Intelligence
Alaa S. Al-Waisy, Shumoos Al-Fahdawi, Mazin Abed Mohammed, Karrar Hameed Abdulkareem, Salama A. Mostafa, Mashael S. Maashi, Muhammad Arif, Begonya Garcia-Zapirain
Summary: The outbreaks of the COVID-19 epidemic have increased the pressure on healthcare and medical systems worldwide. Chest radiography imaging has been shown to be an effective screening technique for diagnosing the COVID-19 epidemic. To reduce pressure on radiologists, a hybrid deep learning framework called COVID-CheXNet has been developed for fast and accurate diagnosis of COVID-19 virus in chest X-ray images. The system achieved high detection accuracy and efficiency, making it a potential tool for real clinical centers.
Article
Chemistry, Analytical
Mazhar Javed Awan, Mohd Shafry Mohd Rahim, Naomie Salim, Amjad Rehman, Begonya Garcia-Zapirain
Summary: This study applies deep learning to automatically segment ACL tears from MRI images. By using the U-Net architecture and semantic segmentation technique, high accuracy segmentation results have been achieved. The method shows promising potential in the field of medical image analysis.
Article
Chemistry, Analytical
Sofia Zahia, Begonya Garcia-Zapirain, Jon Anakabe, Joan Ander, Oscar Jossa Bastidas, Alberto Loizate Totoricaguena
Summary: This study presents a comparative analysis of three different 3D scanning modalities for acquiring 3D meshes of stoma barrier rings from ostomized patients. The results show that the low-cost Structure Sensor structured light 3D sensor has great potential for such applications.
Article
Dermatology
S. Oukil, R. Kasmi, K. Mokrani, B. Garcia-Zapirain
Summary: The study proposes a novel algorithm for discriminating melanoma and benign skin lesions based on color and texture features, achieving high sensitivity (99.25%), specificity (99.58%), and accuracy (99.51%) on the PH2 dataset, outperforming existing methods.
SKIN RESEARCH AND TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Danyal Maheshwari, Daniel Sierra-Sosa, Begonya Garcia-Zapirain
Summary: This article presents the application of a Variational Quantum Classifier (VQC) for binary classification. The study demonstrates the effectiveness of amplitude encoding in enhancing prediction accuracy when applying the VQC method, and achieves high accuracies on various datasets.
Article
Health Care Sciences & Services
Hanane Allioui, Mazin Abed Mohammed, Narjes Benameur, Belal Al-Khateeb, Karrar Hameed Abdulkareem, Begonya Garcia-Zapirain, Robertas Damasevicius, Rytis Maskeliunas
Summary: In this study, a new mask extraction method based on multi-agent deep reinforcement learning (DRL) was introduced and applied to the diagnosis of COVID-19. Experimental validation showed that the method can accurately extract masks of COVID-19 infected areas and achieved good results in pathogenic diagnostic tests and time saving.
JOURNAL OF PERSONALIZED MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Georgina Curto, Mario Fernando Jojoa Acosta, Flavio Comim, Begona Garcia-Zapirain
Summary: This article discusses bias in artificial intelligence, particularly against the poor. The results of the study demonstrate the existence of bias and provide relevant data. Additionally, the evidence shows that bias has important implications for human development in developing countries.
Article
Multidisciplinary Sciences
Zabit Hameed, Begonya Garcia-Zapirain, Jose Javier Aguirre, Mario Arturo Isaza-Ruget
Summary: This paper proposes a deep learning approach for automatic classification of breast cancer microscopy images, achieving good results. The study found that the performance was better on the original dataset, and stain normalization techniques could not surpass the results of the original dataset.
SCIENTIFIC REPORTS
(2022)
Review
Environmental Sciences
S. Shapoval, Merce Gimeno-Santos, Amaia Mendez Zorrilla, Begona Garcia-Zapirain, Myriam Guerra-Balic, Sara Signo-Miguel, Olga Bruna-Rabassa
Summary: This study aims to analyze the solutions found in executive function (EF) training for adults with intellectual disability (ID) in recent years, evaluate them with key parameters, and identify the features and issues in the further development of their system. The results revealed that planning and decision-making were the most frequently mentioned EFs in the analyzed studies, followed by working memory and social cognition, behavioral regulation, flexibility, and inhibition capacity. The trend analysis showed improvements in EFs, although not significant. This study provides significant insights into the creation of support systems and program execution.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Review
Computer Science, Information Systems
Cyrille YetuYetu Kesiku, Andrea Chaves-Villota, Begonya Garcia-Zapirain
Summary: This article discusses the importance of biomedical literature classification and related problems. By analyzing a large number of literature, different classification methods and challenges are understood. The study found that data-centric challenges and data quality challenges are the key issues currently faced in biomedical text classification.
Article
Biology
Mazin Abed Mohammed, Abdullah Lakhan, Karrar Hameed Abdulkareem Begona Garcia-Zapirain, Begona Garcia-Zapirain
Summary: Recently, there has been an increasing ratio of cancer diseases among patients with many cases reported in different clinical hospitals. This paper explores different types of cancer by analyzing, classifying, and processing multi-omics dataset in a fog cloud network. The study proposes new hybrid cancer detection schemes based on SARSA on-policy and multi-omics workload learning.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Genetics & Heredity
Ali Raza, Furqan Rustam, Hafeez Ur Rehman Siddiqui, Isabel de la Torre Diez, Begona Garcia-Zapirain, Ernesto Lee, Imran Ashraf
Summary: This study focuses on predicting genetic disorders using artificial intelligence-based methods. It proposes a novel feature engineering approach and classifier chain approach, and evaluates the performance using multiple evaluation metrics. Results show that extreme gradient boosting (XGB) outperforms state-of-the-art approaches in terms of both performance and computational complexity.
Article
Computer Science, Information Systems
Oscar Jossa-Bastidas, Ainhoa Osa Sanchez, Leire Bravo-Lamas, Begonya Garcia-Zapirain
Summary: Gluten is a natural complex protein found in various cereal grains and can cause immune system attacks in individuals with celiac disease. There are multiple methods for detecting gluten, but this study focuses on developing a novel IoT system using Near-infrared spectroscopy technology and AI algorithms. The results showed high accuracy in predicting the presence or absence of gluten in flour samples.
Review
Medicine, General & Internal
Mazin Abed Mohammed, Karrar Hameed Abdulkareem, Ahmed M. Dinar, Begonya Garcia Zapirain
Summary: This research reviews and evaluates relevant scientific studies on deep learning (DL) models in the omics field, demonstrating their potential and identifying key challenges. The literature survey includes clinical applications, datasets, and highlights the difficulties faced by researchers. Using a systematic approach and specified criteria, 65 articles were selected from four search engines, covering clinical applications, review publications, and comparative analysis guidelines. Obstacles related to DL, preprocessing procedures, datasets, model validation, and testbed applications were identified in the study. The research provides valuable insights and guidance for practitioners in omics data analysis using DL.
