检索结果-内蒙古大学图书馆

1st International Conference in Advanced Innovation on Smart City, ICAISC 2023

作者： Wibowo, Ripto Mukti Fakieh, Bahjat Ramzan, Muhammad Sher Alzahrani, Ahmed Saeed Siddiqui, Muazzam Alzahrani, Bander King Abdulaziz University Faculty of Computing and Information Technology Department of Information Systems Saudi Arabia King Abdulaziz University Faculty of Computing and Information Technology Department of Computer Science Saudi Arabia Indiana University Luddy School of Informatics Computing and Engineering Bloomington United States

ISBN: (纸本)9781665472753

Several regional head elections had to be postponed due to the pandemic, including in Indonesia because of the COVID-19 pandemic. Several big cities in Indonesia are of concern because of their large population and GDP. This study conducts analysis and testing of datasets taken from Open Data in a city in Indonesia. In addition to conducting research on regional head elections, we also present information on voters from the category of kids with disabilities. The steps used in this research are using regional mapping data of the city of Surabaya in the Election of the Regional Head. Download the data or dataset for the Regional Head Election ampersand Categories of kids with disabilities. Based on the dataset voters from the category of children with disabilities are more than 5 *** this research, we use Python to process our datasets & Big Data technology. Data cleaning or cleansing, Exploratory Data Analysis, and Empirical Cumulative Distribution Functions (ECDF) in python are also needed. Result from ECDF chart with steady increase (increment of 0.1). The highest variance value is in Electoral District 5 = 6.090 and the lowest value is in Electoral District 4 = 0.90. The result of Open Data is graphical data visualization and candidate scores to help as an alternative for the 2024 Regional Head Election and the Category of kids with disabilities. © 2023 IEEE.

关键词： Open Data

来源：评论

学校读者我要写书评

暂无评论

Exploitation of a Latent Mechanism in Graph Contrastive Learning: Representation Scattering 38

Exploitation of a Latent Mechanism in Graph Contrastive Lear...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： He, Dongxiao Shan, Lianze Zhao, Jitao Zhang, Hengrui Wang, Zhen Zhang, Weixiong College of Intelligence and Computing Tianjin University Tianjin China Department of Computer Science University of Illinois at Chicago ChicagoIL United States School of Cybersecurity Northwestern Polytechnical University Xi'an China Department of Computing Department of Health Technology and Informatics The Hong Kong Polytechnic University Kowloon Hong Kong

Graph Contrastive Learning (GCL) has emerged as a powerful approach for generating graph representations without the need for manual annotation. Most advanced GCL methods fall into three main frameworks: node discrimination, group discrimination, and bootstrapping schemes, all of which achieve comparable performance. However, the underlying mechanisms and factors that contribute to their effectiveness are not yet fully understood. In this paper, we revisit these frameworks and reveal a common mechanism-representation scattering-that significantly enhances their performance. Our discovery highlights an essential feature of GCL and unifies these seemingly disparate methods under the concept of representation scattering. To leverage this insight, we introduce Scattering Graph Representation Learning (SGRL), a novel framework that incorporates a new representation scattering mechanism designed to enhance representation diversity through a center-away strategy. Additionally, consider the interconnected nature of graphs, we develop a topology-based constraint mechanism that integrates graph structural properties with representation scattering to prevent excessive scattering. We extensively evaluate SGRL across various downstream tasks on benchmark datasets, demonstrating its efficacy and superiority over existing GCL methods. Our findings underscore the significance of representation scattering in GCL and provide a structured framework for harnessing this mechanism to advance graph representation learning. The code of SGRL is at https://***/hedongxiao-tju/SGRL. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Languages Given by Finite Automata over the Unary Alphabet 43

Languages Given by Finite Automata over the Unary Alphabet

引用

43rd IARCS Annual Conference on Foundations of Software Technology and Theoretical Computer Science, FSTTCS 2023

作者： Czerwiski, Wojciech Dbski, Maciej Gogasz, Tomasz Hoi, Gordon Jain, Sanjay Skrzypczak, Micha Stephan, Frank Tan, Christopher Warsaw Poland Faculty of Mathematics Institute of Informatics Informatics and Mechanics School of Informatics and IT School of Informatics and IT Temasek Polytechnic Singapore Singapore School of Computing National University of Singapore Singapore University of Warsaw Poland National University of Singapore Singapore Department of Mathematics National University of Singapore Singapore

ISBN: (纸本)9783959773041

This paper studies the complexity of operations on finite automata and the complexity of their decision problems when the alphabet is unary and n the number of states of the finite automata considered. The following main results are obtained: (1) Equality and inclusion of NFAs can be decided within time 2O((n log n)1/3);previous upper bound 2O((n log n)1/2) was by Chrobak (1986) via DFA conversion. (2) The state complexity of operations of UFAs (unambiguous finite automata) increases for complementation and union at most by quasipolynomial;however, for concatenation of two n-state UFAs, the worst case is an UFA of at least 2O(n1/6) states. Previously the upper bounds for complementation and union were exponential-type and this lower bound for concatenation is new. © 2023 Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing. All rights reserved.

关键词： Finite automata

来源：评论

学校读者我要写书评

暂无评论

The Accuracy Paradox in RLHF: When Better Reward Models Don't Yield Better Language Models

The Accuracy Paradox in RLHF: When Better Reward Models Don'...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Chen, Yanjun Zhu, Dawei Sun, Yirong Chen, Xinghao Zhang, Wei Shen, Xiaoyu Department of Computing The Hong Kong Polytechnic University Hong Kong Saarland University Saarland Informatics Germany Digital Twin Institute Eastern Institute of Technology Ningbo China

ISBN: (纸本)9798891761643

Reinforcement Learning from Human Feedback significantly enhances Natural Language Processing by aligning language models with human expectations. A critical factor in this alignment is the strength of reward models used during training. This study explores whether stronger reward models invariably lead to better language models. In this paper, through experiments on relevance, factuality, and completeness tasks using the QA-FEEDBACK dataset and reward models based on Longformer, we uncover a surprising paradox: language models trained with moderately accurate reward models outperform those guided by highly accurate ones. This challenges the widely held belief that stronger reward models always lead to better language models, and opens up new avenues for future research into the key factors driving model performance and how to choose the most suitable reward models. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Cutting-edge approaches in intrusion detection systems: a systematic review of deep learning, reinforcement learning, and ensemble techniques

引用

Iran Journal of Computer Science 2025年第2期8卷 303-333页

作者： Kalpani, Nethma Rodrigo, Nureka Seneviratne, Dilmi Ariyadasa, Subhash Senanayake, Janaka Department of Computer Science and Informatics Uva Wellassa University Badulla Sri Lanka School of Computing Engineering and Technology Robert Gordon University Aberdeen AB10 7GJ United Kingdom

The growing number of networked devices and complex network infrastructures necessitates robust network security measures. Network intrusion detection systems are crucial for identifying and mitigating malicious activities within network environments. Traditional intrusion detection systems (IDS) often struggle to adapt to new and evolving threats. To address these limitations, researchers are increasingly turning to advanced methodologies such as deep learning (DL), reinforcement learning (RL), and ensemble learning (EL), which offer enhanced capabilities in detecting sophisticated and previously unseen attacks. This systematic literature review (SLR) evaluates 33 technical studies from 2020 to 2024 on the application of DL, RL, and EL in IDS. The study reviews the DL and RL methods used in IDS, along with detection methods that employ EL techniques, to combine the strengths of these approaches. It highlights the advantages, disadvantages, and applicability of the proposed techniques, while also suggesting potential improvements. The aim of this SLR is to provide researchers and practitioners with a comprehensive understanding of current IDS methodologies leveraging DL, RL, and EL, enabling the development of more effective and resilient network security solutions. In addition, this review identifies gaps in the existing literature and outlines potential future research directions. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Deep learning Ensemble learning Intrusion detection systems Network security Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Evaluation of a Contactless Accelerometer Sensor System for Heart Rate Monitoring During Sleep

Evaluation of a Contactless Accelerometer Sensor System for ...

引用

International Conference on Applications in Electronics Pervading Industry, Environment and Society, APPLEPIES 2023

作者： Boiko, Andrei Gaiduk, Maksym Madrid, Natividad Martínez Seepold, Ralf Ubiquitous Computing Lab Department of Computer Science HTWG Konstanz—University of Applied Sciences Konstanz78462 Germany School of Informatics Reutlingen University Reutlingen72762 Germany

ISBN: (纸本)9783031481208

The monitoring of a patient's heart rate (HR) is critical in the diagnosis of diseases. In the detection of sleep disorders, it also plays an important role. Several techniques have been proposed, including using sensors to record physiological signals that are automatically examined and analysed. This work aims to evaluate using a contactless HR monitoring system based on an accelerometer sensor during sleep. For this purpose, the oscillations caused by chest movements during heart contractions are recorded by an installation mounted under the bed mattress. The processing algorithm presented in this paper filters the signals and determines the HR. As a result, an average error of about 5 bpm has been documented, i.e., the system can be considered to be used for the forecasted domain. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Accelerometers

来源：评论

学校读者我要写书评

暂无评论

Society 5.0-Inspired Digitalization Framework for Resilient and Sustainable Agriculture 3rd

Society 5.0-Inspired Digitalization Framework for Resilient ...

引用

3rd International Conference on Society 5.0, 2023

作者： Tombe, Ronald Smuts, Hanlie Future Africa University of Pretoria South Africa Computing Sciences Kisii University Kenya Department of Informatics University of Pretoria South Africa

This research paper proposes a digitalization framework based on Society 5.0 principles for promoting resilient and sustainable agricultural value chains in the context of climate change. Climate change is affecting the productivity and sustainability of agricultural systems and threatening food security in many parts of the world. Digitalization has the potential to enhance the resilience of agricultural value chains to climate change by improving efficiency, promoting sustainability, and reducing vulnerability to climate risks. This study reviews the literature to investigate the potential benefits and challenges of Society 5.0-inspired digitalization for agricultural value chains in the context of resilience and sustainability. Further, this study establishes critical design requirements for digitalization, which inform the development of a theoretical framework for Society 5.0-inspired digitalization framework in realizing resilient and sustainable agricultural value chains. © 2023, EasyChair. All rights reserved.

关键词： Food supply

来源：评论

学校读者我要写书评

暂无评论

The Integration of Deep Learning Techniques and Big Data Analytics for Improved Breast Cancer Diagnosis and Treatment: A Systematic Review

The Integration of Deep Learning Techniques and Big Data Ana...

引用

Information and Communication Technology for Development for Africa (ICT4DA), International Conference on

作者： Hambisa Mitiku Gebre Getachew Mamo Wegari Department of Information Science Faculty of Computing and Informatics Jimma University Jimma Ethiopia Department of Information Technology Faculty of Computing and Informatics Jimma University Jimma Ethiopia

ISBN: (数字)9798331527792

ISBN: (纸本)9798331527808

Background:- Breast cancer is one of the most common and lethal diseases in the world. Traditional breast cancer diagnostic and prognosis procedures usually involve significant human talent and can be time-consuming and subjective, leading to potential errors and treatment delays. Two recent technical advancements, deep learning, and big data analytics are promising to improve breast cancer diagnosis and therapy. As a result, this systematic review aims to examine specific papers that describe deep learning techniques and big data analytics in the context of breast cancer diagnosis and ***: Using Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA), peer-reviewed articles published in the English language from January 2020 to August 2023 were selected from electronic databases such as PubMed, ACM, Digital Library, and Science Direct, as well as citations and manual searches. This review paper takes into account papers about deep learning algorithms, big data analytics, the efficacy of deep learning and big data analytics, as well as the problems and limitations of merging deep learning methods with big data analytics. Articles that were not original or in English were not ***: - Ten articles were identified for this review. The finding showed that deep-learning techniques play a great role in analyzing vast datasets to identify malignant cells or tumors, aiding radiologists in accurate diagnoses and improving patient outcomes. Big data analytics in breast cancer diagnosis and treatment can improve accuracy, efficiency, and patient-centered care. Deep learning techniques are utilized with big data, enhancing screening test accuracy and guiding diagnostic procedures, especially for image-based scanning that leads to early breast cancer identification, improving patient outcomes, and potentially enhancing diagnostics. Obtaining huge volumes of high-quality data for training deep learning models is a challenge and limita

关键词： Deep learning Training Accuracy Merging Medical treatment Manuals Big Data Breast cancer Prognostics and health management Tumors

来源：评论

学校读者我要写书评

暂无评论

Analytical Model for Underwater Wireless Sensor Network Energy Consumption Reduction

引用

Computers, Materials & Continua 2022年第7期72卷 1611-1626页

作者： Huma Hasan Rizvi Sadiq Ali Khan Rabia Noor Enam Kashif Nisar Muhammad Reazul Haque Computer Science Department UBIT Karachi UniversityKarachi75270Pakistan Sir Syed University of Engineering and Technology Karachi75300Pakistan Faculty of Computing and Informatics University Malaysia SabahJalan UMS88400Kota KinabaluSabahMalaysia Faculty of Computing&Informatics Multimedia UniversityPersiaran Multimedia63100CyberjayaSelangorMalaysia

In an Underwater Wireless Sensor Network(UWSN),extreme energy loss is carried out by the early expiration of sensor nodes and causes a reduction in efficiency in the submerged acoustic sensor *** based on clustering strategies,instead of each node sending information by itself,utilize cluster heads to collect information inside the clusters for forwarding collective information to *** can effectively minimize the total energy loss during *** environment of UWSN is 3D architecture-based and follows a complex hierarchical clustering strategy involving its most effecting unique parameters such as propagation delay and limited transmission *** base clustering strategy works in rounds,where each round comprises three fundamental stages:cluster head selection,grouping or node association,and data aggregation followed by forwarding data to the *** UWSN,the energy consumed during the formation of clusters has been considered casually or completely evaded in the previous *** this paper,the cluster head setup period has been considered the main contributor to extra energy utilizer.A numerical channel model is proposed to compute extra *** is performed by using a UWSN broad *** results have shown that extra maximum energy consumption is approximately 12.9 percent of the system total energy consumed in information transmissions.

关键词： Analytical model submerged acoustic sensor systems energy loss round base clustering UWSN extra energy

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Decision Trees, Random Forest, and k-Nearest Neighbors in Predictive Analytics for Orange Telecom's Customer Complaint Data 2

Comparative Analysis of Decision Trees, Random Forest, and k...

引用

2nd International Engineering Conference on Electrical, Energy, and Artificial Intelligence, EICEEAI 2023

作者： Dayyeh, Rayan Alsawareah, Waed Kasasbeh, Bassam Qaddoura, Raneem Kamal, Sinan School of Computing and Informatics Al Hussein Technical University Amman Jordan Data and Artificial Intelligence Department Orange Jordan Amman Jordan

ISBN: (纸本)9798350373363

The study examines the application of machine learning algorithms, namely Decision Trees, Random Forest, and k-Nearest Neighbors, for improving predictive analytics in customer complaint management at Orange Telecom. An analysis is conducted to evaluate the accuracy, efficiency, and scalability of these algorithms in this specific context. Emphasis is placed on thorough data preparation and algorithm assessment, using established performance metrics. Findings provide valuable insights into the handling of complex customer data by these algorithms, underscoring the importance of detailed data preparation for predictive analysis in the telecommunications sector. © 2023 IEEE.

关键词： Nearest neighbor search

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：