检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Lin, Mingquan Xiao, Yunyun Hou, Bojian Wanyan, Tingyi Sharma, Mohit Manoj Wang, Zhangyang Wang, Fei Van Tassel, Sarah Peng, Yifan Population Health Sciences Weill Cornell Medicine New YorkNY United States Department of Biostatistics Epidemiology and Informatics University of Pennsylvania United States Electrical and Computer Engineering The University of Texas Austin United States

In the United States, primary open-angle glaucoma (POAG) is the leading cause of blindness, especially among African American and Hispanic individuals. Deep learning has been widely used to detect POAG using fundus images as its performance is comparable to or even surpasses diagnosis by clinicians. However, human bias in clinical diagnosis may be reflected and amplified in the widely-used deep learning models, thus impacting their performance. Biases may cause (1) underdiagnosis, increasing the risks of delayed or inadequate treatment, and (2) overdiagnosis, which may increase individuals’ stress, fear, well-being, and unnecessary/costly treatment. In this study, we examined the underdiagnosis and overdiagnosis when applying deep learning in POAG detection based on the Ocular Hypertension Treatment Study (OHTS) from 22 centers across 16 states in the United States. Our results show that the widely-used deep learning model can underdiagnose or overdiagnose under-served populations. The most underdiagnosed group is female younger ( © 2023, CC0.

关键词： Diagnosis

SAPIENT: Mastering Multi-turn Conversational Recommendation with Strategic Planning and Monte Carlo Tree Search

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Du, Hanwen Peng, Bo Ning, Xia Department of Computer Science and Engineering The Ohio State University United States Division of Medicinal Chemistry and Pharmacognosy College of Pharmacy The Ohio State University United States Department of Biomedical Informatics The Ohio State University United States Translational Data Analytics Institute The Ohio State University United States

Conversational Recommender Systems (CRS) proactively engage users in interactive dialogues to elicit user preferences and provide personalized recommendations. Existing methods train Reinforcement Learning (RL)-based agent with greedy action selection or sampling strategy, and may suffer from suboptimal conversational planning. To address this, we present a novel Monte Carlo Tree Search (MCTS)-based CRS framework SAPIENT. SAPIENT consists of a conversational agent (S-agent) and a conversational planner (S-planner). S-planner builds a conversational search tree with MCTS based on the initial actions proposed by S-agent to find conversation plans. The best conversation plans from S-planner are used to guide the training of S-agent, creating a self-training loop where S-agent can iteratively improve its capability for conversational planning. Furthermore, we propose an efficient variant SAPIENT-e for trade-off between training efficiency and performance. Extensive experiments on four benchmark datasets validate the effectiveness of our approach, showing that SAPIENT outperforms the state-of-the-art baselines. © 2024, CC BY.

关键词： Chatbots

Implementation of AI-Based Support Vector Machine Algorithm for Fraud Detection in Digital Payment Systems

学校读者我要写书评

暂无评论

Implementation of AI-Based Support Vector Machine Algorithm ...

International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)

作者： Edy Prayitno Rikie Kartadie Uning Lestari Munengsih Sari Bunga Accounting Information Systems Faculty of Information Technology Universitas Teknologi Digital Indonesia Yogyakarta Indonesia Computer Engineering Study Program Faculty of Information Technology Universitas Teknologi Digital Indonesia Yogyakarta Indonesia Informatics Study Program Faculty of Science and Information Technology Universitas Akprind Indonesia Yogyakarta Indonesia Department of Computer Engineering Smart City Information Systems Politeknik Negeri Indramayu Indramayu Indonesia

ISBN: (数字)9798331519643

ISBN: (纸本)9798331519650

The increasing use of digital payment systems has led to a rise in fraudulent activities, presenting a significant challenge in ensuring secure transactions. This research focuses on implementing the Support Vector Machine (SVM) algorithm with a Radial Basis Function (RBF) kernel to detect fraud in digital payment systems. One of the main challenges addressed in this study is the severe class imbalance in the dataset, where fraudulent transactions account for only 0.17% of total transactions. To overcome this, the SMOTE (Synthetic Minority Over-sampling Technique) method was applied to balance the dataset, allowing the model to better recognize fraudulent patterns. The results indicate that the SVM model achieved an accuracy of 99.93%, with a precision of 86.23% and a recall of 75.51%. These results demonstrate that SVM, combined with SMOTE and RBF kernel, is highly effective in detecting fraudulent transactions while minimizing false positives. This research provides a strong foundation for improving fraud detection models in the context of digital payment systems, offering enhanced security and trust for users. Further research could explore hybrid models and real-time data analysis to improve performance.

关键词： Support vector machines Seminars Adaptation models Accuracy Data models Real-time systems Fraud Security Kernel Intelligent systems

Augmented reality for children as a learning media for introduction to historical site

学校读者我要写书评

暂无评论

AIP Conference Proceedings 2024年第1期3116卷

作者： Z. E. Ferdi Fauzan Putra Hamidillah Ajie Claudia Shindi Mersa Ali Idrus Department of Informatics and Computer Engineering Education Universitas Negeri Jakarta Jakarta Indonesia

This research aimed to produce an augmented reality Android-based for historical site named Pesanggrahan Putri Tujuh, which is suitable to be used as an learning media for children. The target of the study was children aged 9-12 years in Dumai city, Riau. The method used in this study is a Multimedia Development Life Cycle (MDLC) with only 5 (five) stages: concept, design, material collecting, assembly, and testing. Feasibility testing of this product used the black box method. The test results showed that product validation by experts got a value of 100% and media experts of 92.6%. In comparison, the product evaluation results by users aged 9-12 years in Dumai City obtained a value of 89.71%. The conclusion is that the product that has been developed is suitable to be used as learning media for the historical site named Pesanggrahan Putri Tujuh.

关键词：

A Streamline-guided De-Homogenization Approach for Structural Design

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Junpeng Westermann, Rüdiger Wu, Jun Computer Graphics and Visualization Department of Informatics Technical University of Munich Garching 85748 Germany Department of Sustainable Design Engineering Delft University of Technology Delft2628 CE Netherlands

We present a novel de-homogenization approach for efficient design of high-resolution load-bearing structures. The proposed approach builds upon a streamline-based parametrization of the design domain, using a set of space-filling and evenly-spaced streamlines in the two mutually orthogonal direction fields that are obtained from homogenization-based topology optimization. Streamlines in these fields are converted into a graph, which is then used to construct a quad-dominant mesh whose edges follow the direction fields. In addition, the edge width is adjusted according to the density and anisotropy of the optimized orthotropic cells. In a number of numerical examples, we demonstrate the mechanical performance and regular appearance of the resulting structural designs, and compare them with those from classic and contemporary approaches. Copyright © 2022, The Authors. All rights reserved.

关键词： Structural design

Automated Fire Hydrant Monitoring and Maintenance using IoT Data Analytics and Naive Bayes Classifier

学校读者我要写书评

暂无评论

Automated Fire Hydrant Monitoring and Maintenance using IoT ...

Distributed Systems, computer Networks and Cybersecurity (ICDSCNC), International Conference on

作者： R. Sarasu V. Nandagopal N V Shibu Suresh Lakshminarayanan T Saju Raj M. Rajmohan Department of Spatial Informatics Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Electrical and Electronics Engineering School of Engineering Mohan Babu University Tirupathi India Department of Computing Technologies School of Computing SRM Institute of Science and Technology Kattankulathur Chennai India Department of Engineering & Electronics University of Technology and Applied Sciences South Al Batina Oman Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai India Department of Electronics and Communication Engineering Hindustan Institute of Technology and Science Chennai India

ISBN: (数字)9798350375442

ISBN: (纸本)9798350375459

This paper introduces a new method for managing fire hydrants that uses a monitoring system based on the Internet of Things (IoT) and a Naive Bayes Classifier (NBC) for predictive maintenance. Water pressure, flow rate, and the physical state of fire hydrants are among the essential factors for the proposed system continually collecting data using an IoT sensor network. The system analyzes real-time data using NBC. These predictive capabilities reduce the potential of hydrant breakdowns during emergencies through preventive maintenance potential. It introduces a reliable IoT infrastructure for fire hydrant monitoring, improving the acquired data's granularity and timeliness. NBC provides a powerful instrument for early problem detection, significantly improving maintenance effectiveness. Since the technology is compatible with existing fire department procedures, it can easily integrate into current operational processes. Better public safety, more efficient use of resources, and more dependable fire hydrants are potential outcomes of responsive and proactive data-driven maintenance techniques. The findings improve overall fire safety and operational efficiency, with a 25% decrease in maintenance expenditures and a 40% increase in issue detection accuracy.

关键词： Accuracy Scalability Urban areas Real-time systems Maintenance Naive Bayes methods Internet of Things Stakeholders Predictive maintenance Risk mitigation

Rethinking Bayesian Learning for Data Analysis: The Art of Prior and Inference in Sparsity-Aware Modeling

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Cheng, Lei Yin, Feng Theodoridis, Sergios Chatzis, Sotirios Chang, Tsung-Hui College of Information Science and Electronic Engineering Zhejiang University Hangzhou China School of Science and Engineering The Chinese University of Hong Kong Shenzhen China Department of Informatics and Telecommunications National and Kapodistrian University of Athens Greece Department of Electronics Systems Aalborg University Denmark Department of Electrical & Computer Engineering and Informatics Cyprus University of Technology Cyprus Cyprus

Sparse modeling for signal processing and machine learning, in general, has been at the focus of scientific research for over two decades. Among others, supervised sparsity-aware learning comprises two major paths paved by: a) discriminative methods that establish direct input-output mapping based on a regularized cost function optimization, and b) generative methods that learn the underlying distributions. The latter, more widely known as Bayesian methods, enable uncertainty evaluation with respect to the performed predictions. Furthermore, they can better exploit related prior information and also, in principle, can naturally introduce robustness into the model, due to their unique capacity to marginalize out uncertainties related to the parameter estimates. Moreover, hyper-parameters (tuning parameters) associated with the adopted priors, which correspond to cost function regularizers, can be learnt via the training data and not via costly cross-validation techniques, which is, in general, the case with the discriminative methods. To implement sparsity-aware learning, the crucial point lies in the choice of the function regularizer for discriminative methods and the choice of the prior distribution for Bayesian learning. Over the last decade or so, due to the intense research on deep learning, emphasis has been put on discriminative techniques. However, a come back of Bayesian methods is taking place that sheds new light on the design of deep neural networks, which also establish firm links with Bayesian models, such as Gaussian processes, and, also, inspire new paths for unsupervised learning, such as Bayesian tensor decomposition. The goal of this article is two-fold. First, to review, in a unified way, some recent advances in incorporating sparsity-promoting priors into three highly popular data modeling/analysis tools, namely deep neural networks, Gaussian processes, and tensor decomposition. Second, to review their associated inference techniques from differ

关键词： Bayesian networks

Real-Time Driver Sleepiness Detection and Classification Using Fusion Deep Learning Algorithm 5th

学校读者我要写书评

暂无评论

Real-Time Driver Sleepiness Detection and Classification Usi...

5th International Conference on Recent Innovations in Computing, ICRIC 2022

作者： Rajawat, Anand Singh Goyal, S.B. Bhaladhare, Pawan Bedi, Pradeep Verma, Chaman Florin-Emilian, Țurcanu Candin, Mihaltan Traian School of Computer Science and Engineering Sandip University Nashik India Faculty of Information Technology City University Malaysia Petaling Jaya Malaysia Galgotias University Greater Noida India Faculty of Informatics University of Eötvös Loránd Budapest1053 Hungary Department of Building Services Faculty of Civil Engineering and Building Services Gheorghe Asachi Technical University of Iasi Jassy700050 Romania Faculty of Building Services Cluj-Napoca Technical University of Cluj-Napoca Cluj-Napoca400114 Romania

ISBN: (纸本)9789811998751

In the past few years, driving while tired has become one of the main causes of car accidents. This kind of dangerous driving can cause serious injuries or even death and huge financial losses. The face, a significant part of the body, can send many different messages. When a person is tired, their facial expressions, like how often they blink their eyes, differ from when they are alert. This is very clear when it comes to drivers. In this study, a fusion deep learning algorithm (FDLA) is used to determine if a driver is tired and how tired they are in real time. This algorithm uses video footage to determine how tired a driver is based on how often they blink and how long their eyes are closed. We are making a new face-tracking algorithm because the ones that are already out there aren’t good enough. This will make face tracking more accurate. Also, a new way to find problems in facility areas has been made, and it is based on seventy essential spots. After that, these parts of the driver’s face can be used to determine how he or she is feeling. When the eyes and lips are used together, a driver care solution can tell if the driver is getting tired. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Learning algorithms

Optimal selection of seed nodes by reducing the influence of common nodes in the influence maximization problem

学校读者我要写书评

暂无评论

Optimal selection of seed nodes by reducing the influence of...

Conference on Information and Knowledge Technology

作者： Farzaneh Kazemzadeh Ali Asghar Safaei Mitra Mirzarezaee Department of Computer Engineering Science and Research branch Islamic Azad University Tehran Iran Department of Medical Informatics Faculty of Medical Sciences Tarbiat Modares University Tehran Iran

ISBN: (纸本)9798350320633

The Internet and social networks play an important role in today's world. Many large companies and organizations use the Internet to manage and implement their activities. Moreover, businesses in these organizations have been significantly expanded through social networks and the appropriate conditions provided by these networks for extensive advertising among a large number of users. Influence maximization (IM) can further spread news and advertisements in social networks by the optimal discovery of a set of k-member of users who have the most diffusion in the social network. Due to its importance, many researchers have investigated this problem and presented algorithms that cannot be used in the real world because they are not optimal and lack proper efficiency. Therefore, this paper reviews and studies the influence maximization problem (IMP) to provide an optimal solution to present an algorithm. The optimal solution is provided by the IBN algorithm at a suitable time. This algorithm prevents the diffusion to be limited among some nodes with common neighbors in a dense part of the network. This method solves the algorithm and selects a set of K members of influential nodes in two steps by scoring for each node and adjusting the scores according to the number of common neighbors between the nodes. The results obtained from this algorithm show that the proposed idea has an optimal solution for this problem, including very high accuracy in selecting the optimal nodes and suitable execution time compared to other algorithms.

关键词： Knowledge engineering Social networking (online) Companies Internet Advertising

Wideband Multi-User MIMO Communications with Frequency Selective RISs: Element Response Modeling and Sum-Rate Maximization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Katsanos, Konstantinos D. Shlezinger, Nir Imani, Mohammadreza F. Alexandropoulos, George C. Department of Informatics and Telecommunications National and Kapodistrian University of Athens Greece School of Electrical and Computer Engineering Ben-Gurion University of the Negev Beer-Sheva Israel School of Electrical Computer and Energy Engineering Arizona State University United States

Reconfigurable Intelligent Surfaces (RISs) are an emerging technology for future wireless communication systems, enabling improved coverage in an energy efficient manner. RISs are usually metasurfaces, constituting of two-dimensional arrangements of metamaterial elements, whose individual response is commonly modeled in the literature as an adjustable phase shifter. However, this model holds only for narrowband communications, and when wideband transmissions are utilized, one has to account for the frequency selectivity of metamaterials, whose response usually follows a Lorentzian-like profile. In this paper, we consider the uplink of a wideband RIS-empowered multi-user Multiple-Input Multiple-Output (MIMO) wireless system with Orthogonal Frequency Division Multiplexing (OFDM) signaling, while accounting for the frequency selectivity of RISs. In particular, we focus on designing the controllable parameters dictating the Lorentzian response of each RIS metamaterial element, in order to maximize the achievable sum rate. We devise a scheme combining block coordinate descent with penalty dual decomposition to tackle the resulting challenging optimization framework. Our simulation results reveal the achievable rates one can achieve using realistically frequency selective RISs in wideband settings, and quantify the performance loss that occurs when using state-of-the-art methods which assume that the RIS elements behave as frequency-flat phase shifters. © 2022, CC BY-NC-ND.

关键词： Broadband networks