检索结果-内蒙古大学图书馆

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Chenhang Wang Yujian Jin Jiayi Liang YuHan Yang Shanling Nie Yang Hai Department of Computer Science and Engineering East China University of Science and Technology Shanghai China Shanghai Key Laboratory of Computer Software Evaluating and Testing Shanghai China Faculty of Engineering The University of Sydney Sydney Australia

Hepatocellular Carcinoma (HCC) holds a record of high incidence and severe global harm. In tasks of liver cancer segmentation based on 3D medical images, the majority of methods have endeavored to enhance the 3D U-net by integrating the latest modules from the field of computer Vision (such as the transformer), often overlooking the distinct characteristics of liver components. We introduced a novel deep learning architecture, Modif-SegUnet, to circumvent this limitation. This architecture extends the U-net model by incorporating a 3D-modifiable attention module, thereby fostering a heightened focus on distinguishing between normal liver sections and lesions. Furthermore, Modif-SegUnet ingeniously amalgamates the 3D-modifiable attention module with the Modif-transformer block, enabling efficient capture of relevant and valuable full-text information in CT images that contain liver or tumor regions. We subjected the proposed Modif-SegUnet to evaluation on the Liver Tumor Segmentation benchmark dataset. Experimental outcomes indicate that our methodology surpasses state-of-the-art approaches in liver tumor segmentation, suggesting potential pathways for advancing the diagnosis and treatment of liver cancer.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A CNN-Based Model for Yield Prediction in MEMS Resonator Manufacturing 36

A CNN-Based Model for Yield Prediction in MEMS Resonator Man...

引用

36th Annual SEMI Advanced Semiconductor Manufacturing Conference, ASMC 2025

作者： Ding, Yikang Wen, Bo Wang, Qi Chen, Chong Lei, Xiangshu Jin, Yufeng Wang, Wei School of Electronic and Computer Engineering Peking University Shenzhen China School of Integrated Circuits Peking University Beijing China School of Software & Microelectronics Peking University Beijing China Faculty of Information and Communication Engineering Dalian University of Technology Dalian China School of Electronic and Computer Engineering School of Integrated Circuits Peking University Beijing China School of Integrated Circuits National Key Laboratory of Advanced Micro and Nano Manufacture Technology Peking University Beijing China

ISBN: (纸本)9798331531850

This work presents a novel method for predicting the yield of Micro-electromechanical Systems (MEMS) resonators by using Convolutional Neural Networks (CNNs) to extract morphological features from cross-sections images of MEMS resonators' beams produced by Deep Reactive-Ion Etching (DRIE). By establishing a model that maps the relationships between the MEMS resonators' operating frequency and the morphological variations in beams induced by process deviations, this method can predict the operating frequency more than 10³ times faster than Finite Element Analysis (FEA) simulation, achieving over 96.7% accuracy, advancing the standardization of MEMS resonators manufacturing flow. © 2025 IEEE.

关键词： convolutional neural networks micro-electromechanical systems yield prediction

来源：评论

学校读者我要写书评

暂无评论

Explaining the Decisions of Deep Learning Models for Load Disaggregation (NILM) Based on XAI

Explaining the Decisions of Deep Learning Models for Load Di...

引用

2022 IEEE Power and Energy Society General Meeting, PESGM 2022

作者： MacHlev, R. Malka, A. Perl, M. Levron, Y. Belikov, J. Technion-Israel Institute of Technology The Andrew and Erna Viterbi Faculty of Electrical & Computer Engineering Haifa3200003 Israel Tallinn University of Technology Department of Software Science Akadeemia tee 15a Tallinn12618 Estonia

ISBN: (数字)9781665408233

ISBN: (纸本)9781665408233

Non-Intrusive Load Monitoring (NILM) techniques estimate the consumption of individual appliances in a household or facility, based on readings of a centralized meter. Lately, deep learning techniques have demonstrated outstanding performance for NILM predictions. Nevertheless, a possible problem is that users and consumers may find it hard to trust the results of such algorithms if they do not fully understand the reasons for their outputs. In this light, this work presents a method that explains and justifies the outputs of NILM convolutional neural network (CNN) classifiers, using Explainable Artificial Intelligence (XAI) techniques. The method operates as follows: a CNN model for NILM is used to estimate which appliances are activated in the system based on the total power consumption. Then, a XAI technique uses this model and its outputs to explain and justify the prediction of this model. Thereby, the NILM CNN classifier outputs are both accurate and are more interpretable, allowing users to make informed and trustworthy decisions. These ideas are demonstrated on the REDD dataset using a convolutional neural network classifier and two state-of-the-art XAI techniques. © 2022 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Faster Edge Coloring by Partition Sieving

arXiv

引用

arXiv 2025年

作者： Akmal, Shyan Koana, Tomohiro INSAIT Sofia University "St. Kliment Ohridski" Bulgaria Technische Universität Berlin Faculty IV Institute of Software Engineering and Theoretical Computer Science Algorithmics and Computational Complexity Germany

In the Edge Coloring problem, we are given an undirected graph G with n vertices and m edges, and are tasked with finding the smallest positive integer k so that the edges of G can be assigned k colors in such a way that no two edges incident to the same vertex are assigned the same color. Edge Coloring is a classic NP-hard problem, and so significant research has gone into designing fast exponential-time algorithms for solving Edge Coloring and its variants exactly. Prior work showed that Edge Coloring can be solved in 2m poly(n) time and polynomial space, and in graphs with average degree d in 2(1−Εd)m poly(n) time and exponential space, where Εd = (1/d)Θ(d3). We present an algorithm that solves Edge Coloring in 2m−3n/5 poly(n) time and polynomial space. Our result is the first algorithm for this problem which simultaneously runs in faster than 2m poly(m) time and uses only polynomial space. In graphs of average degree d, our algorithm runs in 2(1−6/(5d))m poly(n) time, which has far better dependence in d than previous results. We also consider a generalization of Edge Coloring called List Edge Coloring, where each edge e in the input graph comes with a list Le ⊆ {1, . . ., k} of colors, and we must determine whether we can assign each edge a color from its list so that no two edges incident to the same vertex receive the same color. We show that this problem can be solved in 2(1−6/(5k))m poly(n) time and polynomial space. The previous best algorithm for List Edge Coloring took 2m poly(n) time and space. Our algorithms are algebraic, and work by constructing a special polynomial P based off the input graph that contains a multilinear monomial (i.e., a monomial where every variable has degree at most one) if and only if the answer to the List Edge Coloring problem on the input graph is YES. We then solve the problem by detecting multilinear monomials in P. Previous work also employed such monomial detection techniques to solve Edge Coloring. We obtain faster algor

关键词： Undirected graphs

来源：评论

学校读者我要写书评

暂无评论

Integrating AI-based adaptive learning into the flipped classroom model to enhance engagement and learning outcomes

引用

computers and Education: Artificial Intelligence 2025年 8卷

作者： Katona, Jozsef Gyonyoru, Klara Ida Katonane Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary Institute of Electronics and Communication Systems Kandó Kálmán Faculty of Electrical Engineering Obuda University Budapest Hungary GAMF Faculty of Engineering and Computer Science John Von Neumann University Kecskemet Hungary Department of Applied Quantitative Methods Faculty of Finance and Accountancy Budapest University of Economics and Business Budapest Hungary Department of Organizational Development and Communication Science Institute of Social Sciences University of Dunaujvaros Dunaujvaros Hungary

This study explores the integration of the AI-powered adaptive learning system within the flipped classroom model and discusses programming education. The increasing demand for the acquisition of programming skills in the digital era generates the development of innovative pedagogical approaches to help students overcome the difficulties in mastering programming concepts. The study investigates how AI-based adaptive feedback influences student engagement, motivation, and learning outcomes. This study was conducted for 13 weeks, with two groups of undergraduate students. In one of the groups, tshe traditional flipped classroom model was applied, while in the other group, AI-driven adaptive learning tools were introduced. In this research methodology, a mixed-methods approach was adopted, integrating quantitative analyses with qualitative data. These showed that the experimental group demonstrated significant improvements in learning results and motivation, thereby realiszing the value of real-time, personalized feedback provided through the AI system. The qualitative findings revealed that with the AI system, there was an increase in student autonomy;thus, there is increased responsibility among students to collaborate in in-class participation, which helps enhance a more dynamic and cooperative learning environment. On the whole, these findings reveal important implications of the integration of AI into flipped classrooms, while at the same time, they reveal how to make the modernization of programming education achieve a better overall experience in student learning. © 2025 The Authors

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Properties of Nonlinear Power System Models that are Symmetrically Configured in the Narrow Sense

Properties of Nonlinear Power System Models that are Symmetr...

引用

IEEE General Meeting Power& Energy Society

作者： Elior Segev Ron Ofir Yoash Levron Juri Belikov Faculty of Electrical & Computer Engineering Technion—Israel Institute of Technology Haifa Israel Department of Software Science Tallinn University of Technology Tallinn Estonia

Symmetry is a fundamental property of three-phase circuits and is widely used in the analysis of large-scale power systems. However, when the system is nonlinear or time-varying, classical symmetry generally does not provide sufficient information. Therefore, in this paper, we explore properties of three-phase circuits with dynamic models that are “simple” in some stronger sense. We define a circuit to be simple if its state and input matrices in the dq0 domain are constant, that is, independent of the reference angle, and show that this property leads to dynamic models which are linear and time-invariant, or at least suitable for low-complexity numerical analysis. We formally define circuits with this property to be Symmetrically Configured in the Narrow Sense (SCNS), and explore their properties. We show, for instance, that SCNS circuits must be symmetrically configured and that their state-space models have several interesting properties and structures. The overall aim is to develop a small toolbox, which may help identify and analyze SCNS circuits in the context of large-scale power systems. We believe this may contribute to the understanding of dynamic phenomena in large-scale systems which are too complex for direct simulation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A neural networks approach for designing compact all-optical photonic crystal based and logic gate

引用

Journal of Optical Communications 2024年第s1期45卷 s2627-s2636页

作者： Parandin, Fariborz Yahya, Salah I. Rezaeenia, Mehdi Askarian, Asghar Roshani, Saeed Roshani, Sobhan Ghadi, Yazeed Yasin Jamshidi, Mohammad Rezaee, Sahar Department of Electrical Engineering Kermanshah Branch Islamic Azad University Kermanshah Iran Department of Communication and Computer Engineering Cihan University Erbil Iraq Department of Software Engineering Faculty of Engineering Koya University KoyaKOY45 Iraq Department of Electrical Engineering Arak Branch Islamic Azad University Arak Iran Software Engineering and Computer Science Department Al Ain University Al Ain United Arab Emirates Hong Kong Department of Physics Kermanshah Branch Islamic Azad University Kermanshah Iran

This paper introduces a new method for creating an all-optical AND gate by utilizing a two-dimensional photonic crystal configuration for the first time. This gate design is intended for applications in optical computing and all-optical logic, offering the potential for rapid computation and parallel processing. The described gate is characterized by its compact dimensions and comprises two inputs and a single output. The high and low logic states are defined based on power values, where logic 0 corresponds to low power and logic 1 corresponds to high power emitted from the light source. To enhance the design process, artificial neural networks (ANNs) are utilized. ANNs offer a powerful tool for optimizing and fine-tuning the photonic crystal structure parameters to achieve the desired logic functionality. With the help of the applied ANNs, the design process is eased and high performance is achieved for the proposed photonic crystal structure. By integrating ANNs into the design process, this research opens up new possibilities for advancing the field of photonic logic circuits. Combining photonic crystals and ANN optimization provides a powerful approach to designing complex and efficient optical computing systems. The results show that the obtained power values are high for 1 logic state and low for the 0 logic state, which verifies the AND gate accuracy table. The achieved accurate results verify the validity of the proposed approach for achieving precise and reliable all-optical logic operations. © 2023 Walter de Gruyter GmbH, Berlin/Boston.

关键词： Photonic crystals

来源：评论

学校读者我要写书评

暂无评论

The Reliability of Information Sources: Student Perceptions and Educational Opportunities

The Reliability of Information Sources: Student Perceptions ...

引用

International IEEE Conference and Workshop: Óbuda on Electrical and Power engineering (CANDO-EPE)

作者： Anita Mihalovicsne Kollar Jozsef Katona Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary Institute of Electronics and Communication Systems Kandó Kálmán Faculty of Electrical Engineering Obuda University Budapest Hungary GAMF Faculty of Engineering and Computer Science John Von Neumann University Kecskemet Hungary Department of Applied Quantitative Methods Faculty of Finance and Accountancy Budapest Business University Budapest Hungary

ISBN: (数字)9798331531119

ISBN: (纸本)9798331531126

The modern digital age demands the skills to be able to assess information. In the current world, access to information and knowledge is abundant from different sources. Indeed, high-quality data are much needed in education and professional development, as the quality and accuracy of information directly influence the process of learning and decision-making. The study aims to investigate the perceptions of computer science students at the University of Dunaújváros related to the credibility of diverse information sources. Within a questionnaire, on a voluntary answer basis, 68 students took part, 59 males and nine females. This work is aimed at assessing the degree to which the surveyed students evaluate information sources based on a five-point Likert scale. The data shows that students find print books and textbooks as the most reliable sources; social media are the least reliable. The study also supports the importance of source credibility evaluation, for which lateral reading techniques and AI-based news source credibility identification systems can be helpful. These findings suggest the necessity of bettering the educational approaches to foster better source criticism skills in the students, especially for online content and social media. The study generally contributes to understanding the information society and helps prepare students to obtain and use reliable and accurate information in their future careers. Using AI-based tools and systems can then further support these educational efforts with robust mechanisms to evaluate information credibility in an ever more complex digital landscape.

关键词： Economics Power engineering Accuracy Social networking (online) engineering profession Education Reliability engineering Information filters Information age Reliability

来源：评论

学校读者我要写书评

暂无评论

A Roadmap for Artificial Intelligence Augmented software Development Life Cycle: Aspects of Knowledge Vaporization

A Roadmap for Artificial Intelligence Augmented Software Dev...

引用

International Conference on Information Technology (ICIT)

作者： Wael Alzyadat Mohammad Al-Madi Mohammad Muhairat Nagham Al-Madi Feras Altarawneh Department of Software Engineering Faculty of Science and IT Alzaytoonah University of Jordan Amman Jordan Computer Science Al-Zaytoonah University of Jordan Amman Jordan Artificial Intelligence Faculty of Science and IT Alzaytoonah University of Jordan Amman Jordan Department of Software Engineering Faculty of Science and IT Al-Zaytoonah University of Jordan Amman Jordan

Artificial Intelligence Augmented software Development Life Cycle is a process for developing software that incorporates AI techniques and technologies. One aspect of AIASDLC is knowledge vaporization, which refers to the process of extracting and encapsulating knowledge from the AI model during development and making it available for use in other parts of the software development life cycle. The advantages of AIASDLC include; improved performance, increased efficiency, better decision-making, better user experience, better scalability, better security and compliance, and knowledge vaporization. Requirement engineering is the process of gathering, analyzing, and specifying the requirements for a software system. The context boundary, domain knowledge, application domain, and system boundary are all important concepts in requirement engineering that help define the scope and constraints of the system being developed. Understanding the context boundary is important because it helps identify the external factors that will affect the requirements of the system. Domain knowledge is the understanding and expertise in a specific field or area that is necessary for effective requirement engineering. The application domain is the specific area or field where the system will be used, and it is important to understand the application domain to gather and document the correct requirements for the system

关键词：

来源：评论

学校读者我要写书评

暂无评论

SYNFER: TOWARDS BOOSTING FACIAL EXPRESSION RECOGNITION WITH SYNTHETIC DATA

arXiv

引用

arXiv 2024年

作者： He, Xilin Luo, Cheng Xian, Xiaole Li, Bing Song, Siyang Khan, Muhammad Haris Xie, Weicheng Shen, Linlin Ge, Zongyuan Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China AIM Lab Faculty of IT Monash University Australia KAUST Saudi Arabia University of Exter United Kingdom MBZUAI United Arab Emirates

Facial expression datasets remain limited in scale due to privacy concerns, the subjectivity of annotations, and the labor-intensive nature of data collection. This limitation poses a significant challenge for developing modern deep learning-based facial expression analysis models, particularly foundation models, that rely on large-scale data for optimal performance. To tackle the overarching and complex challenge, we introduce SynFER (Synthesis of Facial Expressions with Refined Control), a novel framework for synthesizing facial expression image data based on high-level textual descriptions as well as more fine-grained and precise control through facial action units. To ensure the quality and reliability of the synthetic data, we propose a semantic guidance technique to steer the generation process and a pseudo-label generator to help rectify the facial expression labels for the synthetic images. To demonstrate the generation fidelity and the effectiveness of the synthetic data from SynFER, we conduct extensive experiments on representation learning using both synthetic data and real-world data. Experiment results validate the efficacy of the proposed approach and the synthetic data. Notably, our approach achieves a 67.23% classification accuracy on AffectNet when training solely with synthetic data equivalent to the AffectNet training set size, which increases to 69.84% when scaling up to five times the original size. Our code will be made publicly available. © 2024, CC BY.

关键词：

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：