检索结果-内蒙古大学图书馆

3rd International Conference on Advancements in Smart Computing and Information Security, ASCIS 2024

作者： Tamang, Parimala Thatal, Annet Gupta, Mousumi Department of Computer Application Sikkim Manipal Institute of Technology Sikkim Manipal University Majhitar Sikkim Gangtok India Al-Fahal University Faridabad India

ISBN: (纸本)9783031862984

Colposcope is a device which generates a colposcopic images of the cervix which is vital for detecting cervical neoplasia. Thus, analysis of these images plays a significant role in identifying and diagnosing cervical abnormalities. In this paper, we present a UNET model for automatic segmentation of normal and abnormal regions from colposcopic images. This will be beneficial for swift analysis and minimizes the subjective biases between clinicians. The segmentation works by pre-processing the images, training the UNET model on the set of annotated dataset and finally applying the trained model to predict the normal and abnormal regions. The whole algorithm is tested on Kaggle and ICAR dataset. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Physics-Informed Neural Networks for One-Dimensional Quantum Well Problems

arXiv

引用

arXiv 2025年

作者： Sarkar, Soumyadip Department of Computer Application Narula Institute of Technology India

We implement physics-informed neural networks (PINNs) to solve the time-independent Schrödinger equation for three canonical one-dimensional quantum potentials: an infinite square well, a finite square well, and a finite barrier. The PINN models incorporate trial wavefunctions that exactly satisfy boundary conditions (Dirichlet zeros at domain boundaries), and they optimize a loss functional combining the PDE residual with a normalization constraint. For the infinite well, the ground-state energy is known (E = π2 in dimensionless units) and held fixed in training, whereas for the finite well and barrier, the eigenenergy is treated as a trainable parameter. We use fully-connected neural networks with smooth activation functions to represent the wavefunction and demonstrate that PINNs can learn the ground-state eigenfunctions and eigenvalues for these quantum systems. The results show that the PINN-predicted wavefunctions closely match analytical solutions or expected behaviors, and the learned eigenenergies converge to known values. We present training logs and convergence of the energy parameter, as well as figures comparing the PINN solutions to exact results. The discussion addresses the performance of PINNs relative to traditional numerical methods, highlighting challenges such as convergence to the correct eigenvalue, sensitivity to initialization, and the difficulty of modeling discontinuous potentials. We also discuss the importance of the normalization term to resolve the scaling ambiguity of the wavefunction. Finally, we conclude that PINNs are a viable approach for quantum eigenvalue problems, and we outline future directions including extensions to higher-dimensional and time-dependent Schrödinger equations. © 2025, CC BY.

关键词： Schrodinger equation

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient Location-Based XML Indexing in Real-Time Wireless Data Broadcasting for IoT Devices

Energy Efficient Location-Based XML Indexing in Real-Time Wi...

引用

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Goel, Vikas Sharma, Ashutosh Singh, Mayank Goyal, Amit Kumar Jain, Vikas Kiet Group of Institutions Dept. of It UP Ghaziabad201206 India Ims University Campus Dept. of It UP Ghaziabad201009 India Gl Bajaj Institute of Technology Dept. of Cse Gr. Noida India Chandigarh University Dept. of Cse Punjab 140413 India C.C.S. University Campus Dept. of Computer Application Scriet Meerut India

ISBN: (纸本)9798331508685

The Internet of Things (IoT) is fast revolutionizing a number of sectors by performing real-time data collection and analysis. Effective data broadcasting protocols become increasingly important as IoT devices. Specifically designed for Internet of Things applications, this paper proposes a comprehensive framework for location-based XML indexing in real-time wireless data broadcasting. To enable effective geographic-based XML data searching, we propose a spatial XML indexing technique. Our proposed broadcasting strategy maximizes network use also. The suggested solution ensures energy efficiency of various IoT devices while addressing the issues of scalability and changing data environments. © 2025 IEEE.

关键词： Energy Efficient IoT Location-Based Search (LBS) Wireless Data Broadcasting XML Indexing

来源：评论

学校读者我要写书评

暂无评论

Leveraging Machine Learning for Early Detection of Cardiovascular Diseases

Leveraging Machine Learning for Early Detection of Cardiovas...

引用

Computational, Communication and Information technology (ICCCIT), International Conference on

作者： Anurag Shrivastava Meenakshi Maindola Rakesh Kumar H Pal Thethi S P Sreeja N Sirisha Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai Tamilnadu Department of Computer Science & Engineering Graphic Era Deemed to be University Dehradun Department of Computer Engineering & Applications GLA University Mathura Lovely Professional University Phagwara India Master of Computer Application New Horizon College of Engineering Bangalore Department of Computer Science & Engineering MLR Institute of Technology Hyderabad Telangana

ISBN: (数字)9798331512965

ISBN: (纸本)9798331512972

In order to enhance patient outcomes, rapid and precise detection approaches are needed for cardiovascular diseases (CVDs), which are among the top causes of death globally. By examining a wide range of demographic and clinical data, this study investigates how machine learning approaches could improve the early diagnosis of CVDs. We explore how many different machine learning methods could identify different heart diseases. Decision trees, support vector machines and more, are included in these algorithms. By using feature selection and optimization methods we want to strengthen these models' prediction abilities and make them more resilient. The dataset used by this research includes all of the medical information of patients as they include all the pertinent biomarkers such as age, blood pressure, cholesterol levels and etc. The results suggest that machine learning models, especially ensemble methods like Random Forest and gradient boosting, are indeed able to predict the risk of CVD better than conventional diagnostic strategies. Results suggest that early risk assessment using machine learning embedded within healthcare processes is a reliable and non-invasive approach. This study has found machine learning to have promise for change and further research into the practical use of these models to revolutionize treatment of cardiovascular diseases.

关键词： Support vector machines Accuracy Computational modeling Machine learning Medical services Predictive models Blood pressure Cardiovascular diseases Risk management Random forests

来源：评论

学校读者我要写书评

暂无评论

KLLMs4Rec: Knowledge graph-enhanced LLMs sentiment extraction for personalized recommendations

引用

Expert Systems with applications 2025年 282卷

作者： Cui, Yachao Wang, Kaiguang Yu, Hongli Guo, Xiaoxu Cao, Han School of Computer Technology and Applications Qinghai University Xining810016 China School of Computer Science Shaanxi Normal University Xi'an710000 China School of Artificial Intelligence Institute Beijing Economic Management Vocational College Beijing100018 China Qinghai Provincial Laboratory for Intelligent Computing and Application Qinghai University Xining810016 China

Recommendation algorithms typically leverage auxiliary information such as user reviews and knowledge graphs to enhance algorithm performance, thereby alleviating data sparsity and cold start issues. Recently, researchers have increasingly employed large language models, which boast powerful natural language understanding capabilities, to further improve recommendation systems. However, these models often suffer from hallucination problems. Moreover, integrating heterogeneous information, such as reviews and knowledge graphs, can introduce new noise, potentially impairing recommendation performance. Knowledge graphs, as tightly organized structured knowledge bases, can assist in addressing the hallucination problem and heterogeneous information fusion problem of LLMs. To effectively address the aforementioned issues, we propose the Knowledge Graph-Enhanced Large Language Model Sentiment Extraction for the Personalized Recommendation Model (KLLMs4Rec). It aims to solve the LLMs hallucination problem and the noise problem caused by the fusion of heterogeneous information in recommender systems, and provide users with more accurate, diverse and novel personalized recommendations. To address the hallucination problem when extracting user sentiments from reviews with LLMs, we designed a knowledge graph-enhanced prompt template. It is worth noting that this scheme also solves the noise issue of heterogeneous information fusion. Additionally, to further expand user preferences extracted from reviews, this paper proposes a new hierarchical sentiment attention graph convolutional network, which utilizes three sentiment weight schemes to propagate user personalized preferences on the knowledge graph. Extensive experiments on the Movielens-20 m, Amazon-book, and Yelp datasets demonstrate that our model surpasses current leading methods while effectively addressing the hallucination problem of LLMs and the noise problem of heterogeneous information fusion. © 2025 Elsevier Ltd

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Constrained Multi-objective Optimization Based on Dynamic Priority and Cooperative Offspring Generation

引用

IEEE Transactions on Evolutionary Computation 2025年

作者： He, Zhihui Wang, Feng Li, Bingdong Zhou, Aimin Wuhan University School of Computer Science Hubei 430072 China East China Normal University Shanghai Frontiers Science Center of Molecule Intelligent Syntheses Shanghai Institute of AI for Education School of Computer Science and Technology Shanghai200062 China Ministry of Education Key Laboratory of Advanced Theory and Application in Statistics and Data Science China

As the number and complexity of constraints in constrained multi-objective optimization problems (CMOPs) increase, the performance of existing constrained multi-objective evolutionary algorithms (CMOEAs) declines significantly. A novel idea is to sequentially address each constraint based on priority, effectively reducing the complexity of CMOPs. However, in these algorithms, the constraint-handling priority is determined statically in the initial stage. This may lead to inappropriate determination of constraint-handling priority since accurately estimating the constraint landscape in the initial stage is quite challenging. Moreover, these algorithms tackle constraints separately, neglecting the potential for inter-constraint cooperation and thus compromising their efficiency in constraint handling. Thus, we propose a constrained multi-objective evolutionary algorithm based on dynamic priority and cooperative offspring generation called DPCMOEA. Firstly, the constraint-handling priority is determined dynamically by the estimated inconsistency degree (EID) between the Pareto fronts of the candidate constraints and the current population. Secondly, computational resources are automatically allocated to each constraint according to EID based constraint relationship analysis. Finally, a new offspring generation strategy based on constraint cooperation is designed to enhance the quality of new solutions. Experimental results on six CMOP test suites demonstrate that DPCMOEA outperforms six state-of-the-art algorithms. © 1997-2012 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Research on high-precision online detection method for surface defects of fuel spheres based on three-dimensional information

引用

Measurement: Journal of the International Measurement Confederation 2025年 253卷

作者： Xu, Jiale Huang, Danping Gan, Fangji Liao, Shipeng Luo, Fan Yu, Shaodong Gao, Xiang High-Speed Machine Vision Laboratory School of Mechanical Engineering Sichuan University of Science & Engineering Yibin 644000 China Sichuan University Chengdu 610041 China Chengdu Computer Application Institute Chinese Academy of Sciences Chengdu 610041 China National Institute of Measurement and Testing Technology Chengdu 610041 China

As a critical material in the nuclear industry, fuel balls must ensure a high degree of surface sealing to prevent radiation contamination from leaking. However, due to the spherical shape, traditional methods face challenges in accurately detecting surface defects. To this end, this paper proposes a defect detection method based on three-dimensional reconstruction of the spheres. The method includes a novel automatic acquisition method for 3D point cloud data from spherical surfaces and the online measurement algorithm for geometric quantities of surface defects of the sphere. By utilizing the acquired 3D point cloud data, the algorithm extracts suspicious defect areas on the sphere's surface and calculates deflection and sliding parameters to improve the 3D reconstruction accuracy and defect detection accuracy for shallow surface defects on the sphere. The experimental results show that the proposed online detection method achieves a detection speed of 1.58 s per sphere, with a defect recognition accuracy of 99.9 % and an mean measurement error of 0.16 mm. This method can rapidly, stably, and reliably measure 3D geometric quantities of shallow surface defects of fuel balls and similar spheres, which is of a significant advancement for the sphere manufacturing industry. © 2025 Elsevier Ltd

关键词： Surface defects

来源：评论

学校读者我要写书评

暂无评论

Computing with Smart Rings: A Systematic Literature Review

arXiv

引用

arXiv 2025年

作者： Wang, Zeyu Yu, Ruotong Wang, Xiangyang Ding, Jiexin Tang, Jiankai Fang, Jun He, Zhe Li, Zhuojun Röddiger, Tobias Xu, Weiye Zhang, Xiyuxing Gao, Huan-Ang Gao, Nan Yu, Chun Shi, Yuanchun Wang, Yuntao Key Laboratory of Pervasive Computing Ministry of Education Department of Computer Science and Technology Tsinghua University China Tsinghua University China University of Washington United States Karlsruhe Institute of Technology Germany Intelligent Computing and Application Laboratory of Qinghai Province Qinghai University China

A smart ring is a wearable electronic device in the form of a ring that incorporates diverse sensors and computing technologies to perform a variety of functions. Designed for use with fingers, smart rings are capable of sensing more subtle and abundant hand movements, thus making them a good platform for interaction. Meanwhile, fingers are abundant with blood vessels and nerve endings and accustomed to wearing rings, providing an ideal site for continuous health monitoring through smart rings, which combine comfort with the ability to capture vital biometric data, making them suitable for all-day wear. We collected in total of 206 smart ring-related publications and conducted a systematic literature review. We provide a taxonomy regarding the sensing and feedback modalities, applications, and phenomena. We review and categorize these literatures into four main areas: (1) interaction - input, (2) interaction - output, (3) passive sensing - in body feature, (4) passive sensing - out body activity. This comprehensive review highlights the current advancements within the field of smart ring and identifies potential areas for future research. © 2025, CC BY.

关键词： Wearable computers

来源：评论

学校读者我要写书评

暂无评论

Innovative anti-phishing framework using machine learning approach with evolutionary optimization to secure multimedia applications

引用

Multimedia Tools and applications 2025年 1-28页

作者： Sharma, Bhawna Khurana, Savita Sharma, Gaurav Rani, Sunita Nandal, Sudesh Kumari Chandi, Pragya Rastogi, Rashi Castilla, Ángel Kuc Goyal, Nitin Department of Computer Science & Engineering JMIT Radaur India Department of Information Technology JMIT Radaur India Department of Computer Science & Engineering and Information Technology BPS Mahila Vishwavidalaya Sonipat India Department of Electronics and Communication Engineering BPS Mahila Vishwavidalaya Sonipat India Department of Computer Science & Engineering University Institute of Engineering and Technology Kurukshetra University Haryana India Department of Computer Application Sir Chhotu Ram Institute of Engineering and Technology CCS University Meerut India Engineering Research & Innovation Group Universidad Europea del Atlántico Santander Spain Department of Project Management Universidad Internacional Iberoamericana Campeche México Department of Project Management Universidad Internacional Iberoamericana Arecibo USA Department of Computer Science & Engineering School of Engineering and Technology Central University of Haryana Mahendergarh India

The transmission of users’ personal data via networks poses an increased risk of being exposed to phishing threats. Phishing URLs hiding malicious web addresses have become a widespread means of compromising user privacy. This paper is to improve the accuracy and efficiency of detecting and defending against phishing attacks in multimedia applications, aiming to safeguard users'personal data from exposure via phishing URLs. The proposed approach can be summarized in three phases. In the first phase, cosine and Jaccard similarities between training and test datasets are calculated to obtain a comprehensive understanding of the similarities of data points. A hybrid similarity score is built by combining these two measures in the second phase of the experiment to effectively blend content-related and quantity-related similarities. The third and final phase involves training a model on XGBoost that uses a hybrid similarity score as an input and then tunning up its hyperparameters using a genetic algorithm with the optimization of classification performance. Finally, the created powerful classifiers are merged to form a stacked ensemble model, increasing the overall accuracy of the classification. The strenuous test evaluation of 11,429 rows, 88 columns, through key metrics, such as Precision, Recall, F-Score, and Accuracy show that the proposed anti-phishing model has outperformance parameters in other models. With this lean approach, accuracy in classification can be improved by the incorporation of similarity-based techniques and machine learning and evolutionary optimization, in which defense can be performed against phishing threats in multimedia environments. The new innovation involved in the proposed work is based on stacking an ensemble of similarity-based techniques with a learning and optimization framework of machines, ensuring robustness toward phishing threat defense.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Real-time Abnormal Address Detection for Mobile Devices in Location-based Services

引用

IEEE Transactions on Mobile Computing 2025年

作者： Hong, Zhiqing Yang, Heng Wang, Haotian Lyu, Wenjun Yang, Yu Wang, Guang Liu, Yunhuai Wang, Yang Zhang, Desheng Rutgers University Department of Computer Science PiscatawayNJ08854 United States Beijing Institute of Computer Technology and Application Beijing100854 China JD Logistics Beijing100176 China Lehigh University BethlehemPA18015 United States Florida State University TallahasseeFL32306 United States Peking University Department of Computer Science and Engineering Beijing100871 China University of Science and Technology of China Suzhou215000 China

An address, a textual description of a geographical location, plays an important role in location-based services such as instant delivery. However, abnormal addresses (i.e., an address without detailed or accurate information) have led to significant costs. In real-world settings, abnormal address detection is not trivial because it needs to be completed in real-time to support massive online queries from mobile devices. In this study, we design FastAddr, a fast abnormal address detection framework, which detects abnormal addresses in real time. FastAddr consists of a novel contrastive address augmentation module and a lightweight multi-head attention model. We further design FastAddr+ to enhance FastAddr by utilizing large-scale spatial entities. A comprehensive three-phase evaluation is conducted. (i) We evaluate FastAddr on a real-world dataset and it yields the average F1 of 85.7% in 0.058 milliseconds, which outperforms the state-of-the-art models by 47.4% with a similar detection time. (ii) An offline A/B test shows that FastAddr outperforms the previous model significantly. (iii) We also conduct an online A/B test to compare FastAddr with the deployed model, which shows an improvement of F1 by more than 20%. Moreover, we conduct two case studies on real industry data, demonstrating both the efficiency and effectiveness of FastAddr. © 2002-2012 IEEE.

关键词： Location

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：