检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Gao, Yibo Zhou, Hangqi Gao, Zheyao Wang, Bomin Gao, Shangqi Wang, Sihan Zhuang, Xiahai School of Data Science Fudan University China Dept. of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Dept. of Oncology University of Cambridge United Kingdom

The pursuit of decision safety in clinical applications highlights the potential of concept-based methods in medical imaging. While these models offer active interpretability, they often suffer from concept leakages, where unintended information within soft concept representations undermines both interpretability and generalizability. Moreover, most concept-based models focus solely on local explanations (instance-level), neglecting the global decision logic (dataset-level). To address these limitations, we propose Concept Rule Learner (CRL), a novel framework to learn Boolean logical rules from binarized visual concepts. CRL employs logical layers to capture concept correlations and extract clinically meaningful rules, thereby providing both local and global interpretability. Experiments on two medical image classification tasks show that CRL achieves competitive performance with existing methods while significantly improving generalizability to out-of-distribution data. The code of our work is available at https://***/obiyoag/crl. © 2025, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Cybernated Watermarking Approach on Image with Cryptographic Encryption 2

Cybernated Watermarking Approach on Image with Cryptographic...

引用

2nd International Conference on computer, Communication and Control, IC4 2024

作者： Sivasankari, R. Veeramakali, T. Mustafa Nawaz, S.M. Prakash, P. Umadevi, P. Alvin Ancy, A. B.S.Abdur Rahman Crescent Institute of Science and Technology Dept. of Computer Science and Engineering India Srm Institute of Science and Technology School of Computing Dept of Data Science and Business Systems Tamilnadu Kattankulathur India Sri Sai Ram Institute of Technology Dept of Computer Science and Engineering India Sri Sai Ram Engineering College Dept of Electronics and Communication Engineering India Kpr Institute of Engineering and Technology Department of S & H Coimbatore India Anna University Dept of Electronics and Communication Engineering Chennai India

ISBN: (纸本)9798350387933

Digital watermarking of interactive media content has become an active experimentation field in recent years. In addition to reviewing some of the methods that have been developed for various media types;a general structure for watermark engrafting and identification is outlined. We highlight a few of these application-oriented differences, including copyright protection, substantiation, tamper identification, and data hiding, as well as the technological and system conditions for different media types such as digital photos, digital files, digital audio, and text. We describe a general structure for digital watermarking-based image copyright protection in this approach. In this paper, we outline the pre-processing and implementation of tide marking technique where Twofish algorithm and double encryption is described. The image processing toolbox is also specified by comparing five different image processing techniques. In addition, the post implementation is carried out with the wavelet toolbox and matlab coder to analyze signal, image properties and generate the efficient code for embedding the text. © 2024 IEEE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Effect of Hamming Distance on Performance of ECOC with Estimated Binary Classifiers

Effect of Hamming Distance on Performance of ECOC with Estim...

引用

2022 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2022

作者： Kumoi, Gendo Yagi, Hideki Kobayashi, Manabu Hirasawa, Shigeichi Waseda University Center for Data Science Tokyo Japan University of Electro-Communications Dept. of Computer & Network Eng. Tokyo Japan

ISBN: (数字)9781665452588

ISBN: (纸本)9781665452588

Error-Correcting Output Coding (ECOC) is a method for constructing a multi-valued classifier using a combination of binary classifiers. The effectiveness of ECOC for multivalued classification problems has been demonstrated by many experimental evaluations. Therefore, classification performance have strongly depended on the data under consideration, and it is not clear what kind of combinations of binary classifiers have good performance. Motivated by this fact, the authors have clarified the best combination of binary classifiers that makes ECOC, assuming a situation in which each binary classifier can estimate the true posterior probability. They also have proposed a total framework for analytical evaluation when a binary classifier outputs an estimated posterior probability that approximates the true posterior probability. These studies established a framework for evaluating the theoretical performance of *** on these findings, this study discusses the theoretical performance of ECOC from the upper bound perspective. The results showed that increasing the Hamming distance between code words can blackuce the error rate. We then evaluate various combinations of binary classifiers based on analytical evaluation. © 2022 IEEE.

关键词： Hamming distance

来源：评论

学校读者我要写书评

暂无评论

Optimizing Production and Inventory Management for Defective Items in Manufacturing Systems 3

Optimizing Production and Inventory Management for Defective...

引用

3rd International Conference on Intelligent Systems, Advanced Computing, and Communication, ISACC 2025

作者： Ahmed, Khandakar Rabbi Hossain, Arafat Asif, MD Amaddus Bepary Mohammad, Mahathir Rahaman, Mustafizur Dewan, MD Arafat Miyan Research Institute International University of Business Agriculture and Technology Dhaka Bangladesh Central Michigan University Dept. of Business Data Analytics Mt Pleasant United States Southern California State University Dept. of Business Administration and Management Los AngelesCA United States Southern California State University Dept. of Business Administration and Management United States Westcliff University Dept. of Cloud Engineering IrvineCA United States Los Angeles City College Dept. of Computer Science Los Angeles United States

ISBN: (纸本)9798331523893

Manufacturing defective items is one of the challenges for production efficiency and inventory control. Conventional approaches like Economic Order Quantity (EOQ) and Just-in-Time (JIT) often fall short of the flexibility needed to respond to quality variations in real time. It promotes defect detection and control by employing machine learning (ML) for inventory management. Four classifiers were tested using the UCI SECOM dataset (Bagging, AdaBoost, Gradient Boosting, and Random Forest) to identify defective items. We used effective preprocessing techniques to boost model performance, such as SMOTE for class balancing and PCA for reducing dimensionality and standardization. The bagging and random forest classifiers produced 96.6% and 96.9% test accuracies, with good precision and recall scores. Within this context, we propose an Adaptive Inventory Management Framework comprising machine learning algorithms incorporating predictive analytics to reduce waste, optimize production scheduling(s), and improve resource allocation. These results are consistent with defining Industry 4.0 strategies and show that ML has the potential to transform defect management and inventory optimization in manufacturing systems. © 2025 IEEE.

关键词： Industry 4.0

来源：评论

学校读者我要写书评

暂无评论

Integrated Framework for Simulated Business Intelligence and Real-Time Analytics to Enhance Agility in Industry 4.0 Decision-Making

Integrated Framework for Simulated Business Intelligence and...

引用

Computational Intelligence Approaches and Applications (ICCIAA), International Conference on

作者： Bilal Ibrahim Sowan Ahmad Altarawneh Mohammad Fasha Dept. of Bus. Intell. & Data Analytics University of Petra Amman Jordan Dept. of Computer Science Brunel University of London London UK

ISBN: (数字)9798331523657

ISBN: (纸本)9798331523664

Agility and real-time responsiveness are critical for achieving competitiveness in Industry 4.0. This study introduces an innovative framework that integrates simulation-driven business intelligence and real-time analytics to enhance decision-making processes in dynamic industrial environments. The proposed framework employs advanced multi-agent systems, operational data stores, and machine learning models to process high-velocity data streams, optimize operational workflows, and provide actionable insights. By adopting a layered architecture for data integration, simulation, and decision support, the framework ensures adaptability and scalability. Experimental results demonstrate significant improvements in processing speed, predictive accuracy, and decision latency when compared to traditional approaches. Applications across sectors such as manufacturing and logistics validate its effectiveness, showcasing its ability to transform raw data into valuable insights for strategic and operational decisions. This work establishes a solid foundation for future research, including blockchain integration for secure decision-making and real-time analytics in smart city ecosystems.

关键词： Smart cities Scalability Decision making Transforms Solids Real-time systems Fourth Industrial Revolution Business intelligence Streams Multi-agent systems

来源：评论

学校读者我要写书评

暂无评论

Automated Detection of Poultry Farms from Aerial Images for Actionable AI System toward Biosecurity Applications

Automated Detection of Poultry Farms from Aerial Images for ...

引用

2023 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2023

作者： Li, Yu Das, Rana Duong, Cuong Lim, Teng-Teeh Haithcoat, Timothy Ersoy, Ilker Shyu, Chi-Ren University of Missouri Department of Electrical Engineering and Computer Science ColumbiaMO United States University of Missouri Biological Engineering ColumbiaMO United States University of Missouri Plant Science and Technology Extension ColumbiaMO United States University of Missouri Institute for Data Science and Informatics Columbia United States University of Missouri Institute for Data Science and Informatics Electrical Eng. and Computer Science Dept. ColumbiaMO United States

ISBN: (纸本)9798350359527

Poultry productions have shifted towards larger farms and often cluster in certain regions. However, many of the smaller farms with a considerable amount of production are not considered concentrated animal feeding operation under regulation in many regions, and therefore are not required to obtain a permit. As a result, information of these operations is not being recorded and monitored, contributing to an increased risk for biosecurity issues. The development in computer vision and availability of high-resolution remote sensing imagery allows the possibility for automatic localization and operation estimation of the unregistered farms, filling in the missing information needed for biosecurity risk management. The farm information extracted from images can be incorporated with other available geospatial data, such as road networks, watersheds, etc., to build a comprehensive knowledge base for actionable plans. In the current study, we present a framework for automated localization, analysis, and visualization of poultry farms from remote sensing images. A robust machine learning method was developed to computationally link the visual features from remote sensing imagery to Missouri Department of Natural Resources Concentrated Animal Feeding Operation data, to identify the animal farm type, location, and potential capacity in the region. The developed process can be performed automatically on images of different timestamps to analyze temporal and spatial development of poultry production facilities of a region. This research contributes to the effective management and regulation of poultry and potentially other animal farms, providing insights into farm characteristics and capacities while facilitating informed decision-making. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

The Classification of Timor Weaving Using Artificial Neural Network 9

The Classification of Timor Weaving Using Artificial Neural ...

引用

9th International Conference on Informatics and Computing, ICIC 2024

作者： Aritonang, Mendarissan Setiawan, Adi Krismiyati Wellem, Theophilus Satya Wacana Christian University Faculty of Information Technology Dept of Computer Science Salatiga Indonesia Satya Wacana Christian University Faculty of Science and Mathematics Dept. Mathematics and Data Science Salatiga Indonesia Satya Wacana Christian University Faculty of Information Technology Dept. Informatics & Computer Engineering Education Salatiga Indonesia Satya Wacana Christian University Department of Informatics Salatiga Indonesia

ISBN: (纸本)9798331517601

The number of woven fabric on the island of Timor is very large and varied, making it difficult to distinguish between types and origins. Many woven fabric motifs appear similar but represent different types. Therefore, this classification was used to perform pattern recognition of woven fabric in the system. The process allowed the system's ability to be tested by applying the pattern recognition algorithm to woven images. In this context, the feature extraction method used was gray level co-occurrence matrix (GLCM), while the method used for classification was artificial neural network (ANN). The results showed an accuracy of 87.5% in the system. Finally, GUI system was developed, enabling tests to be performed on woven images for the identification of Timor Weaving image patterns. © 2024 IEEE.

关键词： Artificial Neural Network Feature Extraction Timor Woven Fabric

来源：评论

学校读者我要写书评

暂无评论

Examination of Phishing Attempts on Web-Based Business Applications and Their Preventions

Examination of Phishing Attempts on Web-Based Business Appli...

引用

2024 International Conference on Communication, computer sciences and Engineering, IC3SE 2024

作者： Shaukat, Shahbaz Rehman, Faisal Chhabra, Gunjan Kaushik, Keshav Khan, Inam Ullah Sharef, Hanan Ali, Muhammad Zahid Ghaffar, Irfan Arshad, Muhammad Naeem Lahore Leads University Dept. Of Computer Science & Information Technology Lahore Pakistan Dept. Statistics & Data Science University Of Mianwali Mianwali Pakistan National University Of Sciences & Tech. Nust Islamabad Pakistan Graphic Era Hill University Department Of Cse Dehradun India Graphic Era Deemed To Be University Uttarakhand Dehradun India University Of Petroleum And Energy Studies School Of Computer Science Dehradun India Isra University Department Of Electronic Engineering Seas Islamabad Pakistan

ISBN: (纸本)9798350366846

Because of developments in the online sector, the occurrence of data breaches in web-based business systems is rising quickly. The attackers employ a range of different methods, including Distributed-Denial-of-Service attacks, Hacking Current Sessions, Reusing of Usernames and Passwords, Injections of Structured Query Language, and Phishing Attacks. Phishing attacks are widely used in the modern age. Phishing attacks are frauds attempt for tricking users into giving up sensitive information. Based on the intruder's intended target, such digital threats can damage specific people or whole industries. Researchers in this field offer a variety of phishing identification methods to stop such threats. Throughout this review paper, we categorize modern phishing identification methods into five main groups, which includes framework based on deep learning algorithms and artificial intelligence methodology based on search engines, methodology based on scanned URLs, methodology based on Whitelisting-Blacklisting Using Pictorial Resemblance as a Basis. Additionally, a statistical assessment of every individual category, giving its advantages and disadvantages. Additionally, this article offers users advice on how to prevent cyber-attacks. © 2024 IEEE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

Comparing nnU-Net and deepflash2 for Histopathological Tumor Segmentation

Comparing nnU-Net and deepflash2 for Histopathological Tumor...

引用

34th Medical Informatics Europe Conference, MIE 2024

作者： Hieber, Daniel Haisch, Nico Grambow, Gregor Holl, Felix Liesche-Starnecker, Friederike Pryss, Rüdiger Schlegel, Jürgen Schobel, Johannes DigiHealth Institute Neu-Ulm University of Applied Sciences Germany Pathology Medical Faculty University of Augsburg Germany Institute of Medical Data Science University Hospital Würzburg Germany Dept. of Computer Science Aalen University of Applied Sciences Germany Technical University of Munich Germany

ISBN: (纸本)9781643685335

Machine Learning (ML) has evolved beyond being a specialized technique exclusively used by computer scientists. Besides the general ease of use, automated pipelines allow for training sophisticated ML models with minimal knowledge of computer science. In recent years, Automated ML (AutoML) frameworks have become serious competitors for specialized ML models and have even been able to outperform the latter for specific tasks. Moreover, this success is not limited to simple tasks but also complex ones, like tumor segmentation in histopathological tissue, a very time-consuming task requiring years of expertise by medical professionals. Regarding medical image segmentation, the leading AutoML frameworks are nnU-Net and deepflash2. In this work, we begin to compare those two frameworks in the area of histopathological image segmentation. This use case proves especially challenging, as tumor and healthy tissue are often not clearly distinguishable by hard borders but rather through heterogeneous transitions. A dataset of 103 whole-slide images from 56 glioblastoma patients was used for the evaluation. Training and evaluation were run on a notebook with consumer hardware, determining the suitability of the frameworks for their application in clinical scenarios rather than high-performance scenarios in research labs. © 2024 The Authors.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Sarcasm Prediction Using Different Learning Approaches on User Behavior and Contextual Evaluation 5

Sarcasm Prediction Using Different Learning Approaches on Us...

引用

5th International Conference on Information Management and Machine Intelligence, ICIMMI 2023

作者： Ahire, Lomesh Kautik Babar, S.D. Mahalle, P.N. Dept of Computer Engineering Smt. Kashibai Navale College of Engineering Savitribai Phule Pune University Pune India Dept of Computer Engineering Savitribai Phule Pune University Pune India Dept. of Artificial Intelligence and Data Science Vishwakarma Institute of Information Technology Pune India

ISBN: (纸本)9798400709418

A large number of people have been using social media platforms extensively to communicate their thoughts and feelings in the recent era of social networking. Both the user base and data volume on social networks are growing quickly every day. Any time an event or activity occurs nearby, nearby individuals express their thoughts and reactions on social media. When a new product is introduced, users on social media platforms also comment on it. Some people express their views or feelings using informal or complex language which makes it difficult to understand for another user. It is challenging to ascertain the true thoughts because different people express their opinions in complex ways. In this study, the various factors that affect these feelings are briefly discussed. In order to identify sarcasm on Twitter, a generic technique is also necessary in addition to the tweet's content. The proposed approach uses contents of tweet in association with important aspects like user behavior and context of tweet. By users' behavior we can identify its influence on other users and context is required to identify user behavior while detecting sarcasm. Proposed approach uses user behavior pattern and personality features along with contextual data. This all information and the already known sarcasm prediction mechanism will help us to set up the generic approach to detect sarcasm on Twitter. © 2023 ACM.

关键词： Behavioral research

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：