检索结果-内蒙古大学图书馆

End-to-end entity extraction from OCRed texts using summarization models

Neural Computing and Applications 2024年第35期36卷 22347-22363页

作者： Villa-García, Pedro A. Alonso-Calvo, Raúl García-Remesal, Miguel Biomedical Informatics Group Department of Artificial Intelligence Universidad Politécnica de Madrid Campus Montegancedo s/n Boadilla del Monte Madrid28660 Spain Biomedical Informatics Group Department of Languages Informatic Systems and Software Engineering Universidad Politécnica de Madrid Campus Montegancedo s/n Boadilla del Monte Madrid28660 Spain

A novel methodology is introduced for extracting entities from noisy scanned documents by using end-to-end data and reformulating the entity extraction task as a text summarization problem. This approach offers two significant advantages over traditional entity extraction methods while maintaining comparable performance. First, it utilizes preexisting data to construct datasets, thereby eliminating the need for labor-intensive annotation procedures. Second, it employs multitask learning, enabling the training of a model via a single dataset. To evaluate our approach against state-of-the-art methods, we adapted three commonly used datasets, namely, Conference on Natural Language Learning (CoNLL++), few-shot named entity recognition (Few-NERD), and WikiNEuRal domain adaptation (WikiNEuRal + DA), to the format required by our methodology. We subsequently fine-tuned four sequence-to-sequence models: text-to-text transfer transformer (T5), fine-tuned language net T5 (FLAN-T5), bidirectional autoregressive transformer (BART), and pretraining with extracted gap sentences for abstractive summarization sequence-to-sequence models (PEGASUS). The results indicate that, in the absence of optical character recognition (OCR) noise, the BART model performs comparably to state-of-the-art methods. Furthermore, the performance degradation was limited to 3.49–5.23% when 39–62% of the sentences contained OCR noise. This performance is significantly superior to that of previous studies, which reported a 10–20% decrease in the F1 score with texts that had a 20% OCR error rate. Our experimental results demonstrate that a single model trained via our methodology can reliably extract entities from noisy OCRed texts, unlike existing state-of-the-art approaches, which require separate models for correcting OCR errors and extracting entities. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

A DSL-based Approach for Code Generation and Navigation Process Management in a Single Page Application 14

A DSL-based Approach for Code Generation and Navigation Proc...

引用

14th International Conference on Emerging Ubiquitous systems and Pervasive Networks / 13th International Conference on Current and Future Trends of Information and Communication Technologies in Healthcare, EUSPN/ICTH 2023

作者： Naimi, Lahbib Abdelmalek, Hamza Jakimi, Abdeslam Software Engineering and Information Systems Engineering Team Department of Informatics Faculty of Sciences Techniques of Errachidia University of Moulay Ismail Errachidia52000 Morocco

Single page applications (SPAs) are web applications that offer a fast and responsive user experience, but also pose some challenges in terms of navigation process management and platform compatibility. In this paper, we propose a DSL-based approach for code generation and management of SPA navigation process, based on the visual vocabulary proposed by Garrett. We use a Domain-Specific Language (DSL) to describe the elements of the SPA, such as pages, connectors, and transitions, in a concise and abstract way. We also propose a metamodel for Garrett vocabulary, which defines the concepts and relationships of the DSL. Then, we apply a Model-Driven engineering (MDE) technique to transform the DSL specification into a platform-specific project architecture and basic code for an SPA. We present a case study of using our DSL to generate an Angular SPA and evaluate the benefits of our approach in terms of productivity, maintainability, and reusability. © 2024 Elsevier B.V.. All rights reserved.

关键词： Reusability

来源：评论

学校读者我要写书评

暂无评论

Privacy-preserving hierarchical federated learning with biosignals to detect drowsiness while driving

引用

Neural Computing and Applications 2024年第32期36卷 20425-20437页

作者： López Bernal, Sergio Hidalgo Rogel, José Manuel Martínez Beltrán, Enrique Tomás Quiles Pérez, Mario Martínez Pérez, Gregorio Huertas Celdrán, Alberto Department of Information and Communications Engineering University of Murcia Campus de Espinardo Murcia30100 Spain Communication Systems Group CSG Department of Informatics IfI University of Zurich UZH ZürichCH-8050 Switzerland

In response to the global safety concern of drowsiness during driving, the European Union enforces that new vehicles must integrate detection systems compliant with the general data protection regulation. To identify drowsiness patterns while preserving drivers’ data privacy, recent literature has combined Federated Learning (FL) with different biosignals, such as facial expressions, heart rate, electroencephalography (EEG), or electrooculography (EOG). However, existing solutions are unsuitable for drowsiness detection where heterogeneous stakeholders want to collaborate at different levels while guaranteeing data privacy. There is a lack of works evaluating the benefits of using Hierarchical FL (HFL) with EEG and EOG biosignals, and comparing HFL over traditional FL and Machine Learning (ML) approaches to detect drowsiness at the wheel while ensuring data confidentiality. Thus, this work proposes a flexible framework for drowsiness identification by using HFL, FL, and ML over EEG and EOG data. To validate the framework, this work defines a scenario of three transportation companies aiming to share data from their drivers without compromising their confidentiality, defining a two-level hierarchical structure. This study presents three incremental Use Cases (UCs) to assess detection performance: UC1) intra-company FL, yielding a 77.3% accuracy while ensuring the privacy of individual drivers’ data;UC2) inter-company FL, achieving 71.7% accuracy for known drivers and 67.1% for new subjects, ensuring data confidentiality between companies but not intra-organization;and UC3) HFL inter-company, which ensured comprehensive data privacy both within and between companies, with an accuracy of 71.9% for training subjects and 65.5% for new subjects. © The Author(s) 2024.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A machine learning approach for skin lesion classification on iOS: implementing and optimizing a convolutional transfer learning model with Create ML

引用

International Journal of Computers and Applications 2024年第8期46卷 666-685页

作者： Benedek Szabo, Aron Katona, Jozsef Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary Department of Instrumentation and Automation Institute of Electronics and Communication Systems Kandó Kálmán Faculty of Electrical Engineering Obuda University Budapest Hungary Department of Informatics GAMF Faculty of Engineering and Computer Science Kecskemét Hungary

The integration of machine learning (ML) into mobile applications presents unique challenges, particularly in resource-constrained environments such as iOS devices. Skin lesion classification is a critical task in dermatology, where accurate and efficient diagnostic tools can significantly aid in early detection of malignant lesions. This study aims to implement a machine learning-based iOS mobile application and develop a binary classification model for skin lesion images to determine whether a lesion is malignant. The research utilized Create ML to develop a convolutional neural network optimized for iOS, employing a transfer learning approach. A logistic regression model was cascaded with the convolutional neural network N to enhance classification accuracy. The model's performance was assessed through various validation metrics, ensuring its robustness and efficiency within the constraints of mobile hardware. A curated dataset from the International Skin Imaging Collaboration archive was used for training and testing. The model achieved an accuracy of 92%, a precision of 90%, a recall of 93%, and an F1-score of 91% in classifying skin lesions. These metrics validate the model's efficacy in identifying malignant lesions. Data curation involved collecting, labeling, and preparing a dataset from publicly available sources, ensuring the inclusion of diagnostically relevant features. The final model was integrated into an iOS application using Core ML and Vision frameworks. The developed application demonstrates reliable performance in classifying skin lesions with high accuracy on iOS devices. The inclusion of comprehensive performance metrics justifies the efficacy of the proposed approach. Future work will explore enhancements in model architecture and object detection capabilities to further improve diagnostic precision and application usability. © 2024 Informa UK Limited, trading as Taylor & Francis group.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Analyzing XSS Attack Information Content on Social Media 6

Analyzing XSS Attack Information Content on Social Media

引用

6th International Conference on Cybernetics and Intelligent System, ICORIS 2024

作者： Hanapi, Khaerunnisa Harlina, Sitti Arifin, Suci Ramadhani Arifin, Arham Oktavianus, Michael Ahyuna Department of Information Systems Dipa Makassar University Makassar Indonesia Department of Informatics Engineering Dipa Makassar University Makassar Indonesia Department of Software Engineering Dipa Makassar University Makassar Indonesia

ISBN: (纸本)9798331539603

Social media has become more than just a communication tool;it's now a rapidly evolving platform for sharing information, including cybersecurity threats like XSS attacks. This study investigates if analyzing collections of tweets can identify those containing XSS attack information. By accurately extracting this information, we can effectively classify tweets as containing or not containing XSS attacks. Our research focuses on evaluating how well feature extraction methods perform in achieving this classification accuracy. As the result, the 2-gram model with TF-IDF Vectorizer might be a more suitable choice with accuracy exceeding 97% and demonstrates better tolerance for overfitting due to the inherent properties of TF-IDF. © 2024 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Quality Inference in Federated Learning with Secure Aggregation

引用

IEEE Transactions on Big Data 2023年第5期9卷 1430-1437页

作者： Pejó, Balázs Biczók, Gergely Budapest University of Technology and Economics CrySyS Lab Department of Networked Systems and Services Faculty of Electrical Engineering and Informatics Budapest1111 Hungary ELKH-BME Information Systems Research Group Budapest1111 Hungary

Federated learning algorithms are developed both for efficiency reasons and to ensure the privacy and confidentiality of personal and business data, respectively. Despite no data being shared explicitly, recent studies showed that the mechanism could still leak sensitive information. Hence, secure aggregation is utilized in many real-world scenarios to prevent attribution to specific participants. In this paper, we focus on the quality (i.e., the ratio of correct labels) of individual training datasets and show that such quality information could be inferred and attributed to specific participants even when secure aggregation is applied. Specifically, through a series of image recognition experiments, we infer the relative quality ordering of participants. Moreover, we apply the inferred quality information to stabilize training performance, measure the individual contribution of participants, and detect misbehavior. © 2015 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Enhancing Classification Algorithm Recommendation in Automated Machine Learning: A Meta-Learning Approach Using Multivariate Sparse group Lasso

引用

Computer Modeling in engineering & Sciences 2025年第2期142卷 1611-1636页

作者： Irfan Khan Xianchao Zhang Ramesh Kumar Ayyasamy Saadat M.Alhashmi Azizur Rahim School of Software Dalian University of TechnologyDalian116620China Faculty of Information and Communication Technology Universiti Tunku Abdul RahmanKampar31900Malaysia College of Computing and Informatics University of SharjahSharjah27272United Arab Emirates Department of Computer Systems Engineering University of Engineering and Applied SciencesSwat19200Pakistan

The rapid growth of machine learning(ML)across fields has intensified the challenge of selecting the right algorithm for specific tasks,known as the Algorithm Selection Problem(ASP).Traditional trial-and-error methods have become impractical due to their resource *** Machine Learning(AutoML)systems automate this process,but often neglect the group structures and sparsity in meta-features,leading to inefficiencies in algorithm recommendations for classification *** paper proposes a meta-learning approach using Multivariate Sparse group Lasso(MSGL)to address these *** method models both within-group and across-group sparsity among meta-features to manage high-dimensional data and reduce multicollinearity across eight meta-feature *** Fast Iterative Shrinkage-Thresholding Algorithm(FISTA)with adaptive restart efficiently solves the non-smooth optimization *** validation on 145 classification datasets with 17 classification algorithms shows that our meta-learning method outperforms four state-of-the-art approaches,achieving 77.18%classification accuracy,86.07%recommendation accuracy and 88.83%normalized discounted cumulative gain.

关键词： Meta-learning machine learning automated machine learning classification meta-features

来源：评论

学校读者我要写书评

暂无评论

Towards Threat Modelling of IoT Context-Sharing Platforms 22

Towards Threat Modelling of IoT Context-Sharing Platforms

引用

22nd International Symposium on Network Computing and Applications, NCA 2024

作者： Goudarzi, Mohammad Shaghaghi, Arash Finn, Simon Stiller, Burkhard Jha, Sanjay School of Computer Science and Engineering UNSW Sydney Australia Australia Communication Systems Group Department of Informatics University of Zurich Switzerland Cisco Australia

ISBN: (纸本)9798331510183

The Internet of Things (IoT) involves complex, interconnected systems and devices that depend on context-sharing platforms for interoperability and information exchange. These platforms are, therefore, critical components of real-world IoT deployments, making their security essential to ensure the resilience and reliability of these "systems of systems." In this paper, we take the first steps toward systematically and comprehensively addressing the security of IoT context-sharing platforms. We propose a framework for threat modelling and security analysis of a generic IoT context-sharing solution, employing the MITRE ATT&CK framework. Through an evaluation of various industry-funded projects and academic research, we identify significant security challenges in the design of IoT context-sharing platforms. Our threat modelling provides an in-depth analysis of the techniques and sub-techniques adversaries may use to exploit these systems, offering valuable insights for future research aimed at developing resilient solutions. Additionally, we have developed an open-source threat analysis tool that incorporates our detailed threat modelling, which can be used to evaluate and enhance the security of existing context-sharing platforms. ©2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

PSCS4CPP: A Generative PSCS Implementation for C++ 1

引用

Model-Driven engineering and software Development - 9th International Conference, MODELSWARD 2021, and 10th International Conference, MODELSWARD 2022, Revised Selected Papers

作者： Hammer, Maximilian Maschotta, Ralph Wichmann, Alexander Jungebloud, Tino Bedini, Francesco Zimmermann, Armin Systems and Software Engineering Group Computer Science and Automation Department Technische Universität Ilmenau Ilmenau Germany

ISBN: (数字)9783031388217

ISBN: (纸本)9783031388200

Since the early 2000s, Model-driven engineering has aimed to accomplish executable UML models. With fUML and PSCS, the Object Management group (OMG) published standardized specifications of precise semantics for certain parts of UML in the form of metamodels, which form an execution environment. A certain characteristic of these environments is that static information about a model is analyzed and evaluated during runtime. With composite structures being a concept for describing structural properties of a model, the majority of execution semantics specified by PSCS concern analysis and processing of such static information about the model’s fine-grained structure. Thus, the PSCS specification appears suitable for a generative realization approach. Using Model-To-Text-Transformation to generate source code, which serves as an input for the actual execution environment, the runtime level of model executions can be relieved by outsourcing analysis and processing of static information to the level of code generation. By inserting this preprocessing step, the performance of the actual model execution at runtime can be improved. This paper introduces an implementation of the PSCS specification for C++ based on code generation using Model-to-Text-Transformation. This also includes a generative approach to realize an extension mechanism of the fUML and PSCS execution environments by introducing user-defined semantic execution strategies. The proposed PSCS implementation was developed as a part of the MDE4CPP project. Moreover, this paper presents a set of test models validating the correct functionality of the implementation as well as a performance benchmark. Finally, analysis results of an application example for the presented realization are evaluated and discussed. © 2023, Springer Nature Switzerland AG.

关键词： Specifications

来源：评论

学校读者我要写书评

暂无评论

Concept and Pictogram-Based User-Interface Design of a Helper Tool for People with Aphasia

引用

Studies in Health Technology and informatics 2023年 301卷 77-82页

作者： Mayer, Peter Werner, Katharina Al-Radhi, Mohammed Csapo, Tamas Gabor Czeba, Bálint Nemeth, Géza Rocha, Ana Patrícia Oliveira, Ilídio C. Silva, Samuel Szeker, Melinda Teixeira, António Panek, Paul Applied Assistive Technologies Group HCI Group TU Wien Vienna Austria Department of Telecommunications and Media Informatics Budapest University of Technology and Economics Budapest Hungary EvoAId Kft. Budapest Hungary Institute of Electronics and Informatics Engineering of Aveiro Department of Electronics Telecommunications and Informatics Intelligent Systems Associate Laboratory University of Aveiro Aveiro Portugal Aphasie Association Budapest Hungary

ISBN: (纸本)9781643683867

Background: Aphasia describes the lack of the already gained ability to use language in a common way. "Language" here covers all variations of forming or understanding messages. Objectives: The APH-Alarm project aims to develop a service concept that provides alternative communication options for people with Aphasia to trigger timely help when needed. It considers that a typical user may not be familiar with modern technologies and offers several simple and intuitive options. Methods: The approach is based on event detection of gestures (during daytime or in bed), movement pattern recognition in bed, and an easy-to-use pictogram-based smartphone app. Results: Agile evaluation of the smartphone app showed a promising outcome. Conclusion: The idea of a versatile and comprehensive solution for aphasic people to easily contact private or public helpers based on their actions or automatic detection is promising and will be further investigated in an upcoming field trial. © 2023 The authors, AIT Austrian Institute of Technology and IOS Press.

关键词： Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：