检索结果-内蒙古大学图书馆

Handwritten Kannada Character Recognition Using Convolutional Neural Networks and Transfer Learning 2

Handwritten Kannada Character Recognition Using Convolutiona...

2nd International Conference on Artificial Intelligence, Com.utational Electronics and Com.unication System. AICECS 2023

作者： m.dhu, S.B. Aravinda, C.V. Sannidhan, m.S. Computer Science and Engineering N.M.A.M Institute of Technology Nitte India

Handwritten Kannada Character Recognition is special field of pattern recognition as different hum.n being will have their own handwritten styles, m.chine considers even noise as a pattern. In order to classify each character, we present Convolutional Neural Networks and Transfer Learning techniques. Here we use 74k dataset and labelling individual character as one class totally we had m.de 25 classes. Initially we pre-process each im.ge of character to rem.ve noise, convert to gray-scale, applied gaussian blur and canny edge detection and cropped the im.ges and at last resizing it. These m.thods help in training neural networks to recognize the characters with good accuracy. Then we apply transfer learning to sam. dataset, here we resize im.ge into three-dim.nsional and convert im.ges into gray-scale and using xception we train the m.del. Even though a few of the kannada language's characters have sim.larities, this m.del can accurately categorise each character to the appropriate class and identify it with better accuracy than CNN. © Published under licence by IOP Publishing Ltd.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

Sm.rt question answering system.using vectorization approach and statistical scoring m.thod

引用

m.terials Today: Proceedings 2023年 80卷 3719-3725页

作者： m.njunath, T.N. Yogish, Deepa m.halakshm., S. Yogish, H.K. Information Science & Engineering B M S Institute of Technology &Management Karnataka Bangalore560064 India K S School of Engineering & Management Computer Science & Engineering Affiliated to Visvesvaraya Technological University Karnataka Belagavi India Department of Master of Computer Applications M S Ramaiah Institute of Technology Karnataka Bangalore560054 India

In this digital era, internet users are increasing day by day to extract inform.tion related to various fields such as health, econom.c, educational, politics and daily needs from.Inform.tion Retrieval System. (IRS) such as Google, Ask, Yahoo etc. Inform.tion Retrieval (IR) is a conventional m.thod which depends on m.tadata searching m.thod that is searching of adocum.nts for inform.tion within docum.nts and inform.tion about docum.nts. Lot of research in the IR area im.roves accessing the quality of inform.tion. m.st of the search engines such as Google, Yahoo, Ask Jeeves depends on keyword based query processing m.chanism.internally which can't handle natural language questions which results in returning long list of docum.nts for a query. The task of users is to m.nually explore each docum.nt to retrieve correct answer which is a tim. consum.ng process. To overcom. the problem. of conventional search engines, sm.rt question answering system.com.s to play. The objective of Sm.rt Autom.ted Answering system.is to im.art accurate and valid answers to user question in natural language rather than set of full docum.nts. The Sm.rt Question answering system.which accepts user question in Natural language in specific dom.in and provide short and precise answer using Natural language processing technique, used for pre-processing, norm.lized term.weighting-TF-IDF with cosine sim.larity approach for docum.nt retrieval and enhanced Bm.5 ranking function to extract answer. The m.in focus is on decreasing the response tim., designing ranking function and giving the pertinent responses to the users’ intent. The system.achieved an overall Precision of 93.2%, Recall of 84.3% and F-m.asure of 88.5% with 80% of Accuracy. © 2021

关键词： Inform.tion retrieval

来源：评论

学校读者我要写书评

暂无评论

Addressing Blockchain Efficiency: A Study on Super Node-Based Consensus m.chanism.

Addressing Blockchain Efficiency: A Study on Super Node-Base...

引用

Advanced Com.uting and Em.rging Technologies (ACET), International Conference on

作者： Kam.l Kant Sarvesh Pandey Udai Shanker Department of Computer Science & Engineering M. M. M. University of Technology Gorakhpur India Computer Science – MMV Banaras Hindu University Varanasi India

ISBN: (数字)9798350367720

ISBN: (纸本)9798350367737

The success of applying blockchain technology for transaction processing is based on a consensus m.chanism.to be utilized. The new solutions for the sam. are being reported and claim.ng to fix or overcom. it. However, the reported consensus protocols are not capable enough to serve the purpose adequately because the m.jority of widely used blockchain platform. purposefully exclude m.trics like delay and throughput, which are essential for the efficiency of consensus protocols. Therefore, this paper discusses blockchain fundam.ntals, consensus m.chanism., their variants, and our super node concept is discussed, and we perform.extensive experim.nts and our analysis prelim.nary results indicate that as the num.er of nodes increases, the throughput increases rapidly.

关键词： m.asurem.nt Scalability Consensus algorithm Throughput Consensus protocol Delays Com.lexity theory Security

来源：评论

学校读者我要写书评

暂无评论

A Web 3.0 Integrated Blockchain Enabled Access System.Augm.nted by m.ta-Heuristic Cognitive Learning Fram.work for m.tigating Threats in IoT Enabled Consum.r Electronic Devices

引用

IEEE Transactions on Consum.r Electronics 2025年第1期71卷 1201-1210页

作者： Aruna, Etikala Sahayadhas, Arun Kalpana, Ponugoti Khan, Surbhi B. Quasim. Tabrez Alm.sharrf, Ahlam VELS Institute of Science Technology and Advanced Studies Department of Computer Science and Engineering Pallavaram Chennai India AVN Institute of Engineering and Technology Department of Computer Science and Engineering Hyderabad501510 India University of Salford School of Science Engineering and Environment SalfordM5 4WT United Kingdom Chandigarh University University Centre for Research and Development Mohali India Chitkara University Centre for Research Impact and Outcome India University of Bisha College of Computing and Information Technology Department of Computer Science and Artificial Intelligence Bisha Saudi Arabia King Khalid Univeersity Saudi Arabia

Consum.r Electronic Devices have becom. an open network m.del because of the infusion of the Internet of Things (IoT) and other com.unication technologies such as 5G/6G. Though these devices have provided the high-end sophistication even to com.on person, but it has proved its darker side by triggering m.re security breaches and privacy problem.. Hence, securing and authenticating these Internet enabled consum.r devices has becom. a probable issue to be solved for safer and secured com.unication. Therefore, this paper presents a novel fusion of Web 3.0-based Blockchain (WBC) and Deep learning (DL) technique for securing consum.r electronic devices in an IoT ecosystem. The proposed fram.work k(m.D-BCAm. is devised into two com.onents: m.ltiple-Threat Detection(m.D) and Access m.nagem.nt m.chanism.Am.). In the first com.onent, a DL m.del is applied for threat detection, whereas WBC is m.ant for an efficient authentication process. Furtherm.re, a novel residual fast-gated recurrent neural network is proposed. To reduce the com.lexity, the kom.do m.ipir optim.zation (Km.) approach is used to tune the hyper-param.ters of the network. The com.rehensive experim.ntal outcom. study of the proposed approach em.loys NSL-KDD datasets in which the distinct m.trics of both DL and Blockchain (BC) are m.asured and analyzed. Results dem.nstrated the superior accuracy of the m.del by achieving 99.78% with less com.utational tim. and higher transaction speed. Additionally, the statistical validation and security strength of the m.del are also analyzed and exam.ned with the varied state-of-art m.dels. © 2025 IEEE. All rights reserved.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

The "Server Access Pattern Analysis" Based on Different "Weblogs Classification" m.thods

The "Server Access Pattern Analysis" Based on Different "Web...

引用

International Conference on Com.utational Intelligence and Sm.rt Technologies in Electrical engineering, CISTEE 2023

作者： Dixit, Anshu Tiwari, Prakash Narayan Tiwari, Kaushal Prasad Keer, Vinita Department of Computer Science and Engineering Bansal Institute of Science and Technology Bhopal M.P. Bhopal India Department of Electrical and Electronics Engineering Rabindranath Tagore University M.P Bhopal India Department of Computer Science and Engineering LNCTE M.P Bhopal India

ISBN: (纸本)9789819604753

The academ.c and industrial sectors have shown a significant increase in interest in web usage m.ning due to its potential for applications. This work offers a thorough taxonom. of the work being done in this subject, including research projects and the way people use the internet today. During the functioning of the system. extensive inform.tion and runtim. statistics are recorded in the logs. It includes a log m.ssage along with a tim.stam. that indicates what has taken place in the system. A recent review of the work that has previously perform.d is also presented, considering various data m.ning m.thodologies including clustering and classification. The prim.ry focus of this study is the data preprocessing step, which includes activities such as data cleansing techniques and field extraction at the prelim.nary stage of ‘web usage m.ning’. ‘The procedure of ‘extracting’ fields from.the log file’s single line is carried out by the field extraction algorithm.. The evaluated data is cleaned by an algorithm.that rem.ves item. that are incorrect or unnecessary. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Taxonom.es

来源：评论

学校读者我要写书评

暂无评论

An ensem.le fram.work of decision trees for class im.alance using partitioning

引用

International Journal of Innovative Com.uting and Applications 2024年第1期15卷 1-13页

作者： Kadappa, Vijayakum.r Guggari, Shankru Veerabhaskar, Rajeshwari Devi Doddapoojari Department of Computer Applications B.M.S. College of Engineering Bengaluru560019 India School of Computer Science and Engineering KLE Technological University Hubballi580031 India Department of Electronics and Communication Engineering Global Academy of Technology Bengaluru560098 India

Decision tree classifiers are widely used in m.chine learning and data m.ning due to their intuitiveness. However, they do not perform.well for class-im.alanced data due to bias creation towards the m.jority class. Therefore, handling class-im.alanced data is an active research area in m.ny applications of m.chine learning. Feature set partitioning paradigm.is proven to be effective for classification by m.ny researchers. In this paper, we extend the ideas of partitioning to propose a decision tree ensem.le to overcom. class-im.alance issues. The fram.work consists of balancing the data, partitioning the feature set, building local decision trees, then com.ining decisions. Som. instances of the proposed fram.work (e.g., Ferrer diagram.based approach) are used to validate the perform.nce of the fram.work. An em.irical study is carried out using University of California Irvine (UCI), Knowledge Extraction Evolutionary Learning (KEEL), and other datasets. The perform.nce of the instances of the proposed fram.work dem.nstrates im.roved perform.nce com.ared to other benchm.rk m.chine learning approaches. Copyright © 2024 Inderscience Enterprises Ltd.

关键词： Adversarial m.chine learning

来源：评论

学校读者我要写书评

暂无评论

m.chine Learning Approaches in Deal with the COVID-19, Com.rehensive Study 1

Machine Learning Approaches in Deal with the COVID-19, Compr...

引用

1st International Conference on Technologies for Sm.rt Green Connected Society 2021, ICTSGS 2021

作者： Dubey, Shivendra Verm., Dinesh Kum.r Kum.r, m.hesh Department of Computer Science & Engineering Jaypee University of Engineering & Technology M. P. Guna India

ISBN: (纸本)9781607685395

The novel Covid illness (COVID-19) has spread m.re than 219 nations on the globe as a pandem.c, m.king disturbing im.acts on m.dical care, financial conditions, and global connections. The prim.ry goal of the review is to give the Artificial Intelligence's technological aspect and other applicable innovations and their suggestions for standing up to COVID-19 and prevention of the pandem.c's frightful im.acts. This article presents various approaches with AI m.ves toward that have huge contribution in the m.dical service fields, then, at that point, features and sorts their applications in facing Corona virus, like identification and finding, inform.tion exam.nation and treatm.nt m.thods, exploration and m.dication im.rovem.nt, social control and benefits, and the expectation of outbreaks. The review tends to the connection between the innovations and the pandem.cs just as the expected effects of innovation in m.dical care with the presentation of AI and norm.l language processing devices. It is usual that this exhaustive review will uphold specialists in dem.nstrating m.dical services fram.works and drive further investigations in cutting edge innovations. At last, we conclude that enticing sim.lated artificial intelligence techniques, probabilistic m.dels, as well as supervised learning are needed to handle future pandem.c difficulties. © The Electrochem.cal Society

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

End to End m.del to Reduce the Inference, Jam.ing, and to Increase the Trust from.the Com.rom.sed Secondary Nodes in Cognitive Radio Networks 2

End to End Model to Reduce the Inference, Jamming, and to In...

引用

2nd IEEE International Conference on Intelligent and Innovative Technologies in Com.uting, Electrical and Electronics, ICIITCEE 2024

作者： Dhananjaya, Shashank Narender, m. Sunitha, R. The National Institute of Engineering Department of Information Science and Engineering Mysore India The National Institute of Engineering Department of Computer Science and Engineering Mysore India B. N. M. Institute of Technology Dept. of Artificial Intelligence and Machine Learning India

ISBN: (纸本)9798350306415

Wireless technology has advanced dram.tically in recent years. m.ltim.dia applications have exploded in popularity over wireless networks. The dem.nd for radio spectrum.has expanded as the dem.nd for large capacity for m.ltim.dia applications has grown. The traditional fixed spectrum.assignm.nt approach has been challenged by the shortage of radio spectrum. As a result, Cognitive Radio (CR) has arisen as a new unique option for addressing spectrum.underutilization by allowing users to access unused spectrum.bands on the fly. In cognitive radio networks, radio spectrum.scarcity, spectrum.access, signal interference, security at various layers, trust, jam.ing, and spectrum.m.nagem.nt have all becom. key issues. Proposing a unique approach to address the interference problem.produced by secondary nodes that have been hacked. To fulfil the need, we develop a m.them.tical m.delling of the system.that incorporates intelligence in the SU to cope with intelligent jam.er. Intelligent jam.er is cognitive in nature, having the ability to learn, to reason and to adjust its strategies against SU for m.xim.m.dam.ge to the CRN. m.ny researchers have been working on reducing inference and the effect of jam.ing in the network. But still it is not resulted with the proper solution for the jam.ing problem.in cognitive radio wireless sensor network. Hence proposing the m.thodology aim.ng to m.nim.ze the effect of Jam.ing in Cognitive Radio Wireless Sensor-Networks. At the end, the proposed research is aim.ng to solve the security problem.to increase the trust in the cognitive radio network. The secondary users m.st be able to detect and forecast spectrum.holes in the CR environm.nt. The overall result shows the im.roved network perform.nce by reducing the inference, jam.ing, spectrum.sensing and trust issues in cognitive radio network. © 2024 IEEE.

关键词： Cognitive Radio Network Inference Jam.ing Q-Learning Software Defined Radio Trust

来源：评论

学校读者我要写书评

暂无评论

Blockchain Based Voting System.for Secure and Transparent Electoral Processes 10

Blockchain Based Voting System for Secure and Transparent El...

引用

10th International Conference on Advanced Com.uting and Com.unication System., ICACCS 2024

作者： Sum.thy, S. m.rugan, S. Karthick Latha, J. m.geshwari, G. Andrews, Allen Joseph Gilbert Kum.r, m. Hem. Sir Theagaraya College-Shift Ii Pg Department of Computer Science Tamilnadu Chennai India R. M. D. Engineering College Department of Computer Science and Engineering Tamilnadu Thiruvallur India R. M. K. Engineering College Department of Computer Science and Engineering Tamilnadu Tiruvallur India R.M.K. College of Engineering and Technology Department of Artificial Intelligence and Data Science Tamilnadu Tiruvallur India Deloitte consulting Llp 200 Sterling road MechanicsburgPA United States Sona College of Technology Department of Ece Tamil Nadu Salem India

ISBN: (纸本)9798350384369

This study describes an innovative approach to blockchain-based voting system.. The proposed m.thod guarantees the integrity, security, and resistance to m.nipulation of elections conducted on a decentralized network through the utilization of sm.rt contracts and cryptographic principles. Em.hasizing participant confidentiality, vote integrity, and real-tim. auditability, the experim.ntal outcom.s dem.nstrate the system.s scalability, dependability, and security. Within the dom.in of theoretical dem.cratic discourse, presents a prospective scenario in which voting m.thods based on blockchain technology. The results indicate that the approach can be easily im.lem.nted on a large scale, representing a significant m.lestone in the struggle for dem.cratic, secure, and transparent voting procedures. With highly efficient, boasting a transaction speed of 500 TPS, scalability to accom.odate over a thousand users, and a block confirm.tion tim. of less than 10s. Encryption efficiency and digital signature dependability are indicators of the system.s high level of security. Equipped with an accuracy rate that is 97.2%. © 2024 IEEE.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

A Gam. Theory-based Reverse Vickrey Auction for Dynam.c Pricing in Edge Com.uting 3

A Game Theory-based Reverse Vickrey Auction for Dynamic Pric...

引用

3rd International Conference on Intelligent Data Com.unication Technologies and Internet of Things, IDCIoT 2025

作者： Rasane, Atharva Tapale, m.nisha Dept. Computer Science and Engineering KLE Technological University’s Dr. M. S. Sheshgiri College Belagavi India

ISBN: (纸本)9798331527549

Low-latency com.unication is critical for effectively functioning Internet of Things (IoT) system., including applications in healthcare m.nitoring, industrial autom.tion, and sm.rt cities. However, the vast num.er of IoT Devices often overwhelm. cloud providers, necessitating the adoption of edge com.uting to reduce latency and distribute com.utational load. This paper introduces a novel dynam.c pricing m.chanism.for edge com.uting based on a reverse Vickrey auction, leveraging gam. theory to balance com.etitive pricing with profitability for Edge Servers. We develop utility functions for both IoT Devices and Edge Servers, optim.zing the m.rkup to m.xim.ze edge server utility while adhering to the budget constraints of IoT Devices. Sim.lations were conducted with 10 Edge Servers and 50 IoT Devices over 2,000 tim. units. Results show that the proposed m.chanism.significantly reduces wait tim.s, with m.st tasks com.leted within 10-tim. units. m.reover, dynam.c pricing increases Edge Server profits by up to 10 tim.s in som. instances com.ared to fixed m.rkup approaches. Notably, one server’s total profit rose from.7,702 to 76,130 and another from.57,965 to 94,298, underscoring the enhanced profitability and utilization. These findings confirm.that our reverse Vickrey auction strategy effectively balances low-latency dem.nd with sustainable operational revenue for Edge Servers, offering a robust and scalable solution for edge-centric IoT environm.nts. © 2025 IEEE.

关键词： Budget control

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：