检索结果-内蒙古大学图书馆

2023 Italia Intelligenza Artificiale - Thematic Workshops, Ital-IA 2023

作者： Dentamaro, Vincenzo Impedovo, Donato Pirlo, Giuseppe Semeraro, Gianfranco Department of Computer Science University of Studies of Bari Via Edoardo Orabona 4 Bari Italy University School for Advanced Studies IUSS Pavia Piazza della Vittoria 15 Pavia Italy

Process mining is a field of research that has gained much attention in recent years because of its ability to analyze and improve processes. Indeed, one of the key aspects of process mining is its ability to predict the activities in the future and the time spent on these activities. In this work is proposed the use of Bidirectional LSTM and Multi-Speed Transformer on a recent dataset called BPIC-2020 related to reimbursement process of the University of Technology of Eidenhoven. Results shows that Multi-Speed Transformer is more capable to performs next activity prediction than the Bi-LSTM. Meanwhile, for the elapsed time prediction, the viceversa is true. © 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Public administration

来源：评论

学校读者我要写书评

暂无评论

MedFS: Pursuing Low Update Overhead via Metadata-Enabled Delta Compression for Log-structured File System on Mobile Device 23

MedFS: Pursuing Low Update Overhead via Metadata-Enabled Del...

引用

23rd USENIX Conference on File and Storage Technologies, FAST 2025

作者： Wu, Chao Ji, Cheng Chang, Li-Pin Zhu, Zongwei Gao, Congming Guo, Weichao Yu, Chao Wang, Yanzhi Department of Computer Science and Engineering Nanjing University of Science and Technology China Department of Computer Science National Yang Ming Chiao Tung University Taiwan Suzhou Institute for Advanced Research University of Science and Technology of China China School of Information Science Xiamen University China Guangdong Oppo Mobile Telecommunications Corp. Ltd. China College of Engineering Northeastern University United States

ISBN: (纸本)9781939133458

The increasing deployment of data-intensive applications on mobile devices poses a formidable challenge in designing flash-based file systems tailored to these needs. studies have shown that adopting delta compression in log-structured file systems is promising for such an environment as it can effectively reduce the write stress and improve flash longevity. Unfortunately, delta compression suffers from large maintenance overhead. While prior works have introduced nonvolatile memory buffer or battery-backed DRAM to mitigate this, they are less appealing for cost-sensitive mobile devices. This paper introduces MedFS, a Metadata-enabled delta compression on log-structured File System for mobile devices, to achieve a good design trade-off in log-structured file systems employing delta compression. Through a comprehensive analysis of mobile applications and file update patterns, we develop delta-inlining technique, which consolidates delta updates within the inline area of the file's inode block. By leveraging the inherent inode structure and automatically flushing dirty inodes to storage, we effectively address the maintenance overhead associated with delta compression. Additionally, we propose a complementary delta maintenance strategy that selectively manages delta chunks in the data area, overcoming the space constraints of the inline area. Experimental results show that MedFS significantly reduces the write traffic by 55.1% on average, leading to the prolonged storage endurance by 122.7% and improved I/O performance by up to 37.3% over existing work. © 2025 FAST. All Rights Reserved.

关键词： Nonvolatile storage

来源：评论

学校读者我要写书评

暂无评论

Extraction of Logo from Real Time Document Images Using Masking and Median Filter Approaches 3

Extraction of Logo from Real Time Document Images Using Mask...

引用

3rd International Conference for Emerging Technology, INCET 2022

作者： Shivakumar, G. Ravikumar, M. Shivaprasad, B.J. Guru, D.S. Kuvempu University Department of Computer Science Karnataka Shivamogga India Srinivasa Institute of Techonology Department of Computer Science and Engineering Karnataka Mangaluru India University of Mysore Department of Studies in Computer Science Manasagangothri Karnataka India

ISBN: (纸本)9781665494991

In this paper, we propose an efficient method for extracting the logo from real-time document images. A logo represent key perspectives and thoughts in a specific item, making it the most important choice for brands with a lot to say. These logos can include line-drawn symbols, shapes, patterns, or illustrations that use non-literal illustrations to communicate brand awareness. Frequently combined with a logotype. The proposed method extract the logo based on median filter and masking techniques. Experimentation is conducted with two stages namely logo extraction with enhancement and logo extraction without enhancement. Experimentation is carried out on our own data set containing 500 document images and have obtained 90% accuracy for logo extraction with enhancement and 86% for without enhancement. © 2022 IEEE.

关键词： Median filters

来源：评论

学校读者我要写书评

暂无评论

Adaptive Configuration for Heterogeneous Participants in Decentralized Federated Learning 42

Adaptive Configuration for Heterogeneous Participants in Dec...

引用

42nd IEEE International Conference on computer Communications, INFOCOM 2023

作者： Liao, Yunming Xu, Yang Xu, Hongli Wang, Lun Qian, Chen University of Science and Technology of China School of Computer Science and Technology China University of Science and Technology of China Suzhou Institute for Advanced Study China University of California Jack Baskin School of Engineering Department of Computer Science and Engineering Santa Cru United States

ISBN: (纸本)9798350334142

Data generated at the network edge can be processed locally by leveraging the paradigm of edge computing (EC). Aided by EC, decentralized federated learning (DFL), which overcomes the single-point-of-failure problem in the parameter server (PS) based federated learning, is becoming a practical and popular approach for machine learning over distributed data. However, DFL faces two critical challenges, i.e., system heterogeneity and statistical heterogeneity introduced by edge devices. To ensure fast convergence with the existence of slow edge devices, we present an efficient DFL method, termed FedHP, which integrates adaptive control of both local updating frequency and network topology to better support the heterogeneous participants. We establish a theoretical relationship between local updating frequency and network topology regarding model training performance and obtain a convergence upper bound. Upon this, we propose an optimization algorithm, that adaptively determines local updating frequencies and constructs the network topology, so as to speed up convergence and improve the model accuracy. Evaluation results show that the proposed FedHP can reduce the completion time by about 51% and improve model accuracy by at least 5% in heterogeneous scenarios, compared with the baselines. © 2023 IEEE.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

How Do Hackathons Foster Creativity? Towards Automated Evaluation of Creativity at Scale 25

How Do Hackathons Foster Creativity? Towards Automated Evalu...

引用

2025 CHI Conference on Human Factors in Computing Systems, CHI 2025

作者： Falk, Jeanette Chen, Yiyi Rafner, Janet Zhang, Mike Bjerva, Johannes Nolte, Alexander Computer Science Aalborg University Copenhagen Denmark Department of Computer Science Aalborg University Copenhagen Denmark Center for Hybrid Intelligence Department of Management Aarhus University Aarhus Denmark Aarhus Institute for Advanced Studies Aarhus University Aarhus Denmark Software Engineering and Technology Eindhoven University of Technology Eindhoven Netherlands

ISBN: (纸本)9798400713941

Hackathons have become popular collaborative events for accelerating the development of creative ideas and prototypes. There are several case studies showcasing creative outcomes across domains such as industry, education, and research. However, there are no large-scale studies on creativity in hackathons which can advance theory on how hackathon formats lead to creative outcomes. We conducted a computational analysis of 193,353 hackathon projects. By operationalizing creativity through usefulness and novelty, we refined our dataset to 10,363 projects, allowing us to analyze how participant characteristics, collaboration patterns, and hackathon setups influence the development of creative projects. The contribution of our paper is twofold: We identified means for organizers to foster creativity in hackathons. We also explore the use of large language models (LLMs) to augment the evaluation of creative outcomes and discuss challenges and opportunities of doing this, which has implications for creativity research at large. © 2025 Copyright held by the owner/author(s).

关键词： Hackathons

来源：评论

学校读者我要写书评

暂无评论

A comparative stratification of Automated Indagation on Apple Disease Using DeepSpectral Generative Adversarial and Densenet Convolutional Neural Network

A comparative stratification of Automated Indagation on Appl...

引用

Advances in Computing, Communication and Applied Informatics (ACCAI), International Conference on

作者： V. Subha K. Kasturi Department of Computer Science Vels Institute of Science Technology & Advanced Studies (VISTAS) Chennai India Department of Information Technology School of Computing Sciences Vels Institute of Science Technology & Advanced Studies (VISTAS) Chennai India

ISBN: (数字)9798350389449

ISBN: (纸本)9798350389456

A prominent and rife among fruit is apple, one of the significant fruits that are a globally established source of human *** growing demand for improved quality of crops and augmenting the resilience of diseases is a field of expertise researched in multiple dimensions. The existing studies in the chosen area of apple fruit disease have incorporated diverse domains such as machine learning and other algorithmic effectuations. However, the persisting glitches in terms of ineffective disease segmentation, less sensitivity, and specificity leading to constraints in enhanced classification results are the impetus to the proposed research. The proposed study provides a juxtaposed scrutinization of two deep-learning-based solutions, such as the Deep Spectral Generative Adversarial Network (DSGANs), to categorize the diseases that are triggered in apples and the DenseNet Recursive Convolutional Neural Network (DNRCNN) for optimized classification. The rendered solutions are effectuated through different phases such as preprocessing, segmentation, feature extraction from the apple fruit images, and algorithm performance metrics. However, the results explicitly indicate that the DenseNet Recursive Convolutional Neural Network (DNRCNN) provides higher stratified throughput than the former due to the influence of optimized feature extraction. The simulations for this study are implemented using Python, and the results are successfully procured.

关键词： Accuracy Biological system modeling Feature extraction Generative adversarial networks Throughput Real-time systems Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

Covertvasion: Depicting threats through covert channels based novel evasive attacks in android

引用

International Journal of Intelligent Networks 2023年第1期4卷 337-348页

作者： Gautam, Sunil Pattani, Ketaki Zuhair, Mohd Rashid, Mamoon Ahmad, Nazir Department of Computer Science and Engineering Institute of Technology Nirma University Ahmedabad Gujarat India Department of Computer Sciences and Engineering Institute of Advanced Research Gujarat Gandhinagar India School of Information Communication and Technology Bahrain Polytechnic Isa Town Bahrain Department of Information System College of Applied Sciences King Khalid University P.O. Box 61913 Muhayel63317 Saudi Arabia

Privacy and security issues concerning mobile devices have substantial consequences for individuals, groups, governments, and businesses. The Android operating system bolsters smartphone data protection by imposing restrictions on app behavior. Nevertheless, attackers conduct systematic resource analyses and divert privacy-sensitive information from plain view. They employ evasive mechanisms to evade system monitoring and create an illusion of benign and non-sensitive communication. Furthermore, covert channels amplify the impact of these malicious activities by facilitating information transfer through non-standard methods. The purpose of this research is to shed light on these novel threats targeting Android systems. The study delves into security and privacy attacks that compromise sensitive user information. The methodology leverages evasion concepts and employs sound-specific covert channel communication, particularly ultrasonic channels. This research work introduces novel evasive attacks, namely Prime-Composite Evasive Information Invasion (PCEII) and File-lock-based Evasive Information Invasion (FEII), both relying on covert channel communication. These unique variants of attacks successfully evade user data within a few milliseconds for both noisy as well as non-noisy environments and do not show any signs of detection by antivirus mechanisms like Anti-Virus Guard (AVG), 360 security, etc. and state-of-the-art tools such as TaintDroid, MockDroid and others. The paper not only assesses their impact on the privacy and security of information but also introduces avenues for their detection and mitigation. © 2023 The Authors

关键词： Android (operating system)

来源：评论

学校读者我要写书评

暂无评论

Ultrahigh-resolution atomic localization via superposition of standing waves

引用

Communications in Theoretical Physics 2025年第3期77卷 57-65页

作者： Muhammad Idrees Ahmed S Hendy Zareen A Khan Institute of Nonlinear Physics and Department of Physics Zhejiang Normal UniversityJinhua 321004China Zhejiang Institute of Photoelectronics and Zhejiang Institute for Advanced Light Source Zhejiang Normal UniversityJinhua 321004China Department of Computational Mathematics and Computer Science Institute of Natural Sciences and MathematicsUral Federal University19 Mira St.Yekaterinburg 620002Russia Department of Mechanics and Mathematics Western Caspian UniversityBaku 1001Azerbaijan Department of Mathematical Sciences College of SciencePrincess Nourah Bint Abdulrahman UniversityPO Box 84428Riyadh 11671Saudi Arabia

In this study we theoretically demonstrate ultrahigh-resolution two-dimensional atomic localization within a three-levelλ-type atomic medium via superposition of asymmetric and symmetric standing wave *** analysis provides an understanding of the precise spatial localization of atomic positions at the atomic level,utilizing advanced theoretical approaches and principles of quantum *** dynamical behavior of a three-level atomic system is thoroughly analyzed using the density matrix formalism within the realm of quantum mechanics.A theoretical approach is constructed to describe the interaction between the system and external fields,specifically a control field and a probe *** absorption spectrum of the probe field is thoroughly examined to clarify the spatial localization of the atom within the proposed configuration.A theoretical investigation found that symmetric and asymmetric superposition phenomena significantly influence the localized peaks within a two-dimensional spatial ***,the emergence of one and two sharp localized peaks was observed within a one-wavelength *** observed notable influences of the intensity of the control field,probe field detuning and decay rates on atomic ***,we have achieved an unprecedented level of ultrahigh resolution and precision in localizing an atom within an area smaller thanλ/35×λ/*** findings hold promise for potential applications in fields such as Bose-Einstein condensation,nanolithography,laser cooling,trapping of neutral atoms and the measurement of center-of-mass wave functions.

关键词： ultrahigh resolution atom localization symmetric and asymmetric superposition probe absorption standing wave fields

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence Technique for Prediction of Carbon Stocks and Uncertainty Estimates in Tropical Forests

引用

SN computer science 2025年第4期6卷 1-21页

作者： Akinbiola, Sehinde Akinsola, Jide Ebenezer Taiwo Ajagbe, Sunday Adeola Salami, Ayobami Olubaju, Ayomide Awotoye, Olusegun Awoleye, Olusesan Micheal Department of Surveying and Geoinformatics First Technical University Ibadan Ibadan 200255 Nigeria Department of Computer Science First Technical University Ibadan Ibadan 200255 Nigeria Department of Computer Engineering First Technical University Ibadan Ibadan 200255 Nigeria Institute of Ecology and Environmental Studies Obafemi Awolowo University Osun State Ile Ife Nigeria African Institute for Science Policy and Innovation Obafemi Awolowo University Ile Ife Nigeria

The lack of a robust national forest inventory in tropical regions remains a major challenge to the assessment of forest productivity and the evaluation of the climate regulation potential of tropical forests. Therefore, many sampling procedures and models exist for estimating carbon stocks;albeit with varying levels of limitations because of their associated errors. Given the development in computing and modeling in environmental science, this study focuses on providing an Artificial Intelligence (AI)-based approach for predicting carbon stocks and uncertainty estimates in tropical forests. In this study, an artificial intelligence (AI) technique was used to model carbon stocks using machine learning (ML) paradigms. Four ML algorithms, namely, Multi-Layer Perceptron (MLP), Support Vector Machines (SVM), Random Forest (RF), and K-Nearest Neighbors (KNN), were considered based on biophysical variables;including tree height and diameter. The model was implemented in the Waikato Environment for Knowledge Analysis and Jupyter (6.1.4) using the Anaconda distribution of Python (3.00). Uncertainty estimates were evaluated using cost performance measures, including Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE), to validate the reliability of the predicted carbon stock values. In addition, a comparative analysis was performed using 8 different performance measures for classification, while error estimation was performed for regression analysis. One of the models produced included the random forest machine learning algorithm with an accuracy of 98.3%, root mean square error of 0.02, and model building time of 0.13 s, which was chosen as the best algorithm for modeling carbon stocks. This study concluded that there is variation in carbon distribution among species which is not proportional to the volume of carbon stocks in the genus. The study recommends for further studies, the application of a multi-criteria decision-making approach in the selection of an opti

关键词： Artificial intelligence Carbon stock Global climate regulation Informatics Machine learning Tropical-forest productivity

来源：评论

学校读者我要写书评

暂无评论

Recent Advancements in Hyperspectral Face Recognition: A Review 13

Recent Advancements in Hyperspectral Face Recognition: A Rev...

引用

13th International Conference on Computing Communication and Networking Technologies, ICCCNT 2022

作者： Das, Samiran Chatterjee, Sujoy JIS University Centre for Data Science JIS Institute for Advanced Studies and Research Kolkata India Amity University Department of Computer Science and Engineering Kolkata India

ISBN: (数字)9781665452625

ISBN: (纸本)9781665452625

The increasing demand for high-accuracy face recognition methods and the rapid advancement in instrumentation has made hyperspectral face recognition a viable alternative to traditional face recognition. The high-security biometric application scenarios, which demand high accuracy of recognition are switching to hyperspectral imaging due to its robustness against spoofing, illumination, pose variation, etc. The hundreds of visible and near infrared (NIR) images contained in a hyperspectral face cube provide sufficient information content, leading to more accurate recognition. However, the gigantic volume of information and the computational challenges call for efficient learning strategies for the recognition task. Researchers have introduced several feature extraction, supervised learning, deep learning, and pattern recognition approaches to carry out face recognition from hyperspectral face cubes. This work summarized the state-of-the-art approaches for hyperspectral face recognition in detail and highlight their pros and cons. The work also discusses future directions and challenges in this domain. © 2022 IEEE.

关键词： Biometrics

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：