检索结果-内蒙古大学图书馆

Blockchain and explainable-AI integrated system for Polycystic Ovary Syndrome (PCOS) detection

PeerJ computer science 2025年 11卷 e2702-e2702页

作者： Jaganathan, Gowthami Natesan, Shanthi Department of Computer Science and Engineering Kongu Engineering College Tamilnadu Erode India

In the modern era of digitalization, integration with blockchain and machine learning (ML) technologies is most important for improving applications in healthcare management and secure prediction analysis of health data. This research aims to develop a novel methodology for securely storing patient medical data and analyzing it for PCOS prediction. The main goals are to leverage Hyperledger Fabric for immutable, private data and to integrate Explainable Artificial Intelligence (XAI) techniques to enhance transparency in decision-making. The innovation of this study is the unique integration of blockchain technology with ML and XAI, solving critical issues of data security and model interpretability in healthcare. With the Caliper tool, the Hyperledger Fabric blockchain’s performance is evaluated and enhanced. The suggested Explainable AI-based blockchain system for Polycystic Ovary Syndrome detection (EAIBS-PCOS) system demonstrates outstanding performance and records 98% accuracy, 100% precision, 98.04% recall, and a resultant F1-score of 99.01%. Such quantitative measures ensure the success of the proposed methodology in delivering dependable and intelligible predictions for PCOS diagnosis, therefore making a great addition to the literature while serving as a solid solution for healthcare applications in the near future. © (2025), (PeerJ Inc.). All rights reserved.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Private Inference in Quantized Models

引用

IEEE Transactions on Information Theory 2025年第5期71卷 3957-3973页

作者： Deng, Zirui Ramkumar, Vinayak Bitar, Rawad Raviv, Netanel Washington University in St. Louis Department of Computer Science and Engineering St. LouisMO63130 United States Tel Aviv University Department of Electrical Engineering-Systems Tel Aviv-Yafo6997801 Israel Institute for Communications Engineering Technical University of Munich Munich80333 Germany

A typical setup in many machine learning scenarios involves a server that holds a model and a user that possesses data, and the challenge is to perform inference while safeguarding the privacy of both parties. Private Inference has been extensively explored in recent years, mainly from a cryptographic standpoint via techniques like homomorphic encryption and multiparty computation. These approaches often come with high computational overhead and may degrade the accuracy of the model. In our work, we take a different approach inspired by the Private Information Retrieval literature. We view private inference as the task of retrieving inner products of parameter vectors with the data, a fundamental operation in many machine learning models. We introduce schemes that enable such retrieval of inner products for models with quantized (i.e., restricted to a finite set) weights;such models are extensively used in practice due to a wide range of benefits. In addition, our schemes uncover a fundamental tradeoff between user and server privacy. Our information-theoretic approach is applicable to a wide range of problems and robust in privacy guarantees for both the user and the server. © 1963-2012 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Enhancing Customer Retention: A Federated Machine Learning Framework for Banking Churn Prediction 3

Enhancing Customer Retention: A Federated Machine Learning F...

引用

3rd IEEE International Conference on Automation and Computation, AUTOCOM 2025

作者： Vanitha, M. Faritha Banu, J. Srm Institute of Science and Technology Faculty of Engineering and Technology Department of Computer Science and Engineering Bharathi Salai Ramapuram Tamil Nadu Chennai India

ISBN: (数字)9798331542375

ISBN: (纸本)9798331542375

Customer attrition has become the significant challenge for the bank, making large volume of customers to migrate to other banks, as the banks keeps providing multiple benefits to the incoming customers. The loss due to migration of the existing customer to the competitive bankers creates the banking churn, means of loss of customer relationship with the bank, and affects the development, business and the profitability of the corresponding bank. It is essential to predict the banking churn with a primary objective of retaining the customers, considered as a critical task. To achieve this objective, banking sectors employ the customer behavioral analysis to determine the rate of customer churn, resulting in incorrect diagnosis of the churn rate. To over this concern, this research manuscript proposes a novel federated Machine Learning (ML) framework for the prediction of the customer churn, directly contributes for the enhancement of the customer retention. This proposed novel framework offers significant advancements for the banking prediction on customer retention, thus provides an accurate tool for the analysis of customer relationships. The proposed prediction framework is analyzed in terms of accuracy, precision, recall, F1 score and the performance is compared with the state of the art prediction methodologies. © 2025 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

DDoS Detection in SDN Environments: Leveraging Machine Learning Algorithms with Local Interpretable Model-Agnostic Explanations (LIME) for Enhanced Security

DDoS Detection in SDN Environments: Leveraging Machine Lear...

引用

International Conference on Business Intelligence and Data Analytics, BIDA 2024

作者： Karna, Siwani Jahanve, Poluru Reddy Yadukul, S. Bhaskaran, Sreebha Vekkot, Susmitha Rajagopal, Shinu M. Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bangalore India Department of Electronics and Communication Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Bangalore India

ISBN: (纸本)9789819777167

SDN, short for Software Defined Networking, represents a network architecture controlled by software applications. Distributed Denial-of-Service (DDoS) attacks create a challenge to these SDN environments and lead to many issues like loss of data and network instability, and takes a lot of time to address requests made by human users. It is very crucial to detect such attacks, for users to seamlessly use the network and take advantage of what is available online. This study demonstrates the significance of Machine Learning algorithms in identifying malicious traffic. Employing algorithms Gradient Boosting Classifier (GBC), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Logistic Regression (LR) proves effective in detecting complex and dynamic attacks. The study utilizes Local Interpretable Model-agnostic Explanation (LIME) to analyze the strength of each feature that contributes towards the prediction of classification. The results highlight the superior performance of the Gradient Boosting Classifier, achieving an accuracy of 99.75% and an F1-score of 99.68% in detecting network anomalies and threats. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

PregAN-NET: Addressing Class Imbalance with GANs in Interpretable Computational Framework for Predicting Safety Profile of Drugs Considering Adverse Reactions During Pregnancy

引用

Journal of Biomedical Informatics 2025年 166卷 104832页

作者： Chaurasia, Anushka Kumar, Deepak Yogita Department of Computer Science and Engineering National Institute of Technology Meghalaya India Department of Computer Engineering National Institute of Technology Kurukshetra India

Adverse Drug Reactions (ADRs) during pregnancy pose significant risks to both the mother and the fetus. Conventional approaches to predict ADR are inadequate due to ethical restrictions that prevent performing medication studies in pregnant women, leading to restricted data samples. Hence, computational techniques have been promising for ADR predictions. However, most of these techniques have focused on the general population and face the challenge of class imbalance and lack of model interpretability. In the present work, an ensemble learning-based PregAN-NET framework has been proposed that addresses the issue of class imbalance by generating synthetic data employing Conditional Tabular Generative Adversarial Network (CTGAN) and integrates neural network and gradient boosting as a Boosted Neural Ensemble (BNE) architecture to predict safe and unsafe drugs considering their adverse reactions during pregnancy. Furthermore, the SHAP method has been employed to enhance the post-hoc interpretability of the BNE architecture by analyzing the contribution of different features towards prediction. The proposed framework has been applied to chemical and biological properties from PubChem and DrugBank, along with class labels from the ADReCS database. CTGAN has been evaluated for data balancing, showing a 2% to 5% performance improvement over SMOTE. The BNE architecture has outperformed six state-of-the-art methods by achieving mean ROC-AUC scores between 77.00% and 90.00% for chemical data, 66.00% and 74.00% for biological data, and 70.00% to 75.00% for combined datasets. Further, the top 20 contributory features in prediction corresponding to the different drug properties have been identified. © 2025 Elsevier Inc.

关键词： Obstetrics

来源：评论

学校读者我要写书评

暂无评论

The Role of Generative AI in Deepfake Detection: A Systematic Literature Review

The Role of Generative AI in Deepfake Detection: A Systemati...

引用

International Conference on Smart Business and technology, ICSBT 2024

作者： Acim, Btissam Kharmoum, Nassim Lagmiri, Souad Najoua Ziti, Soumia Department of Computer Science Intelligent Processing & Security of Systems Faculty of Sciences Mohammed V University in Rabat Rabat Morocco Rabat Morocco IT Networking Security and Management Laboratory Higher Institute of Management Administration and Computer Engineering Rabat Morocco

ISBN: (纸本)9783031866975

There is a possibility that deepfakes will increase a great deal of false information in a very realistic way, but it seems that today such technology appears to be extremely dangerous. While generative Artificial Intelligence (AI) does expand possibilities and give an opportunity to achieve terrific outcomes and has innumerable potential uses. This systematic literature review focuses on the role of generative AI in the improvement of deepfake detection technologies, providing a comparison between different methods and assessing their effectiveness. By analyzing recent studies, it depicts the techniques employed most frequently in deepfake detection, such as the ones concentrating on the intricacies between the visual and auditory elements, the temporal domains, as well as any minor signs of alteration. Due to the fact that generative AI is rapidly improving, the methods that are used to detect its use are often inadequate, which calls for more studies on the subject matter. Besides, by combining audio and visual elements when performing the analysis, calling it as a method that can be useful in focusing the detection growth potential with diverse and unstructured data. Also, this study evaluates the advantages and limitations of the techniques existing today and demonstrates their relative effectiveness in controlled and real conditions. Clarifying that measures aimed at detecting AI generated content should be preoccupied with a technique that is able to keep pace with changes in generative AI. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Optimize Decision Model for Urban Land Development Using Disc Fermatean Fuzzy Information With Z-Numbers

引用

IEEE Access 2025年 13卷 68370-68383页

作者： Xia, Wei Liu, Yang Pi, Ya China University of Geosciences School of Geography and Information Engineering Hubei Wuhan430078 China South China University of Technology School of Computer Science and Engineering Guangdong Guangzhou510641 China Wuhan Business University School of Foreign Languages Hubei Wuhan430056 China

Urban land development choices manifest complex, ambiguous situations where multiple criteria are involved, making it hard for traditional evaluation methods to address them effectively. The present study introduces an advanced methodological view called the Disc Fermatean Fuzzy Z-Number (D-FFZN) method developed to enhance the accuracy and reliability of urban planning decisions by skillfully addressing uncertainties. The D-FFZN approach was applied in the selection of optimal land development alternatives in a case study through newly formulated aggregation operators based on the weighted averaging and weighted geometric methods. To confirm the method’s performance, a comparative analysis was undertaken between the D-FFZN method and a more renowned multi-criteria decision-making technique, the Combined Compromise Solution (CoCoSo). The findings suggested that D-FFZN offered a more comprehensive and balanced assessment framework that leads to more accurate and reliable decision outcomes. Sensitivity analysis demonstrated that small changes in the parameter value produced minor alterations in the ranking order of alternatives but the D-FFZN model provided a robust optimal solution regardless, thus proving its sturdiness. These findings indicate that the D-FFZN technique has a potential to substantially improve urban land-use planning by equipping stakeholders, policymakers, and urban planners with invaluable guidance in selecting optimal and sustainable development alternatives. © 2013 IEEE.

关键词： Sensitivity analysis

来源：评论

学校读者我要写书评

暂无评论

Spatial Augmented Reality Assistance System with Sensor-Based Activity Recognition at Cleaning Activities 7

Spatial Augmented Reality Assistance System with Sensor-Base...

引用

7th IEEE International Conference on Artificial Intelligence and eXtended and Virtual Reality, AIxVR 2025

作者： Fukasawa, Mana Mikuriya, Masayuki Ogino, Fumitoshi Nakayama, Yu Tokyo University of Agriculture and Technology Department of Electrical Engineering and Computer Science Tokyo Japan Duskin Co. Ltd. Osaka Japan

ISBN: (纸本)9798331521578

Spatial Augmented Reality (SAR) is a promising option for creating interactive AR in real environments. Projection-based SAR provides visual support with digital images projected on physical surfaces. There was no prior study on a system to visually support the learning of full-body movements using tools. Therefore, in this paper we propose a SAR assistance system with sensor-based activity recognition. We implemented a SAR system for supporting floor cleaning activities as a typical use case. The participant's activities are identified in real-time using fixed 2D-LiDARs and an IMU sensor mounted on a cleaning tool. The short-focus floor projector updates the projected image based on the detected activities to support and encourage the participant. The contribution of the proposed idea is to integrate SAR and real-time sensor-based activity recognition. This paper reports a comprehensive evaluation of the proposed SAR system, including the effectiveness of the visualizations. The experimental results suggest that the proposed system is effective as a training system, as it not only motivates users to train, but also encourages them to correct their posture. © 2025 IEEE.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

DR-Minerva: A Multimodal Language Model Based on Minerva for Diagnostic Information Retrieval 23rd

DR-Minerva: A Multimodal Language Model Based on Minerva f...

引用

23rd International Conference of the Italian Association for Artificial Intelligence, AIxIA 2024

作者： Siragusa, Irene Contino, Salvatore Pirrone, Roberto Department of Engineering University of Palermo Sicily Palermo90128 Italy Department of Computer Science IT University of Copenhagen København S2300 Denmark

ISBN: (纸本)9783031806063

This paper illustrates the development of Minerva Diagnostic Retriever (DR-Minerva), a Visual Language Model specialized in the medical domain. Prompted using a textual input with the patient’s information along with a CT or MR scan, the model provides information about the body part and the scanning modality of the given image. The model relies on the Flamingo architecture, which is well known for its good in-context and few-shot learning capabilities, and it encodes textual data using Minerva, a novel Large Language Model trained on English and Italian data. Model performances are improved via fine-tuning the aforementioned model, and using external knowledge by means of a Retrieval Augmented Generation approach. At inference time, the model is injected with the retrieved examples in form of in-context learning. The authors developed a rearranged version of the MedPix® multi-modal medical dataset, that was used for both the development and the test of the model as long as for retrieval. A detailed description of the system is reported along with the experimental results that are discussed in thoroughly. Dataset and models used are available on GitHub (https://***/CHILab1/MedPix-2.0.). © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Analysis of Microgrids Electric Vehicles Charging Stations with Hybrid Energy Storage System Algorithm with AI

Analysis of Microgrids Electric Vehicles Charging Stations w...

引用

International Conference on Computational Intelligence and Smart Technologies in Electrical engineering, CISTEE 2023

作者： Tiwari, Prakash Narayan Tiwari, Kaushal Prasad Dixit, Anshu Bohre, Aashish Kumar Shah, Nand Lal Gupta, Sanjeev Kumar Department of Electrical and Electronics Engineering Rabindranath Tagore University Bhopal India Department of Computer Science and Engineering Bansal Institute of Science and Technology Bhopal India Department of Electrical and Electronics Engineering NIT Durgapur West Bengal Durgapur India

ISBN: (纸本)9789819604753

Renewable energy sources, such as hydropower, solar power, and wind power, have the capacity to efficiently supply their respective portions of the world’s energy needs. Since then, the use of renewable energy in electric cars (EVs) has garnered a lot of attention. Significant advancements in lithium-ion batteries have expedited the development of electric vehicles. The issue is that as the number of EVs increases, so does the need for electricity, increasing the burden on the power system. This promotes research into alternative and sustainable energy sources for recharging electric vehicles. In this study, a solar energy system is used to build an EV charging station. The charging station uses many connectors to supply a DC bus with a constant voltage. Power balance and constant current/constant voltage charging are the foundations of the charging controllers. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Hybrid vehicles

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：