检索结果-内蒙古大学图书馆

28th International computer science and Engineering Conference, ICSEC 2024

作者： Janprasit, Siwach Punkong, Narong Ratanavilisagul, Chiabwoot Kosolsombat, Somkiat Faculty of Applied Science Department of Computer and Information Science Bangkok Thailand Digital Technology for Business Faculty of Management Science Kanchanaburi Thailand Faculty of Applied Science Department of Computer and Information Sciences Bangkok Thailand Thammasat University Data Science and Innovation College of Interdisciplinary Studies Thailand

ISBN: (纸本)9798350366860

handwritten digit recognition is a crucial task in various fields such as postal mail sorting, bank check processing, and digitizing handwritten documents. This research aims to compare the effectiveness of using Convolutional Neural Network (CNN) for feature extraction combined with Support Vector Machine (SVM) for classification, against the traditional method of using Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT) for feature extraction. Additionally, Particle Swarm Optimization (PSO) is applied to optimize SVM parameters to further enhance model performance. The experiments are conducted on two datasets: The MNIST dataset and an individual handwriting dataset from 50 participants. The results indicate that the proposed method of using CNN for feature extraction and PSO-optimized SVM achieves superior accuracy compared to traditional DWT and DCT methods. Specifically, the proposed method achieved an accuracy of 99.05% on the MNIST dataset and 67.79% on the individual handwriting dataset, demonstrating its robustness and effectiveness in handwritten digit recognition. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Cybersecurity: A Multilayered Approach to Phishing Website Detection Using Machine Learning

Enhancing Cybersecurity: A Multilayered Approach to Phishing...

引用

2023 IEEE International Conference on Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering, RMKMATE 2023

作者： Lakshmi Narayanaa, T. Skandarsini, R.R. Ida, S. Jhansi Dept. of Computer Science and Business Systems RMK Engineering College Thiruvallur India

ISBN: (纸本)9798350305708

This paper employs machine learning techniques to combat the escalating threat of phishing attacks in the digital realm. The research builds a predictive model capable of differentiating between phishing and legitimate websites by examining a broad range of information retrieved from website URLs, including address bar-based properties, domain characteristics, and webpage content. To ascertain their effectiveness in this task, six well-known machine learning algorithms-Decision Tree, Random Forest, XGBoost, Deep Learning, Autoencoder Neural Network, and Support Vector Machines-are rigorously investigated. Notably, the Multilayer Perceptrons algorithm emerges as the standout performer, achieving an 86.4% accuracy in identifying phishing websites. This endeavor not only advances the field of cybersecurity but also empowers individuals to proactively safeguard themselves against the pervasive threat of phishing attacks in an increasingly interconnected digital landscape. © 2023 IEEE.

关键词： Cybersecurity Dataset Machine learning model Phishing attacks prediction

来源：评论

学校读者我要写书评

暂无评论

Optimizing Artificial Neural Networks Trough Weight Adjustments 28th

Optimizing Artificial Neural Networks Trough Weight Adjustme...

引用

28th International Conference on Knowledge Based and Intelligent information and Engineering systems, KES 2024

作者： Akber, Syed Muhammad Abrar Szczesna, Agnieszka Kazmi, Sadia Nishat Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice44-100 Poland Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice44-100 Poland

Longer training times pose a significant challenge in artificial neural networks (ANNs) as it may leads to increasing the computational costs and decreasing the effectiveness of the model. Therefore, it is imperative to reduce training times in ANNs to enhance the computational efficiency. The initialization of the weights between the layers in ANN plays a vital role in reducing training times. Appropriate weight initialization can help the network converge faster during the training by providing an optimum starting point for the network. Therefore, weight initialization techniques are essential for efficient training of ANNs. This paper revisits and implements different popular weight initialization techniques in ANNs and analyzes their impact on training time. Specifically, this paper implements Gaussian-based, Kaming-based, and Xavier-based weight initiation atop a popular DNN-based network. The experiments are conducted by employing a well-known dataset. The results show that the scenario when no weight initiation is applied consumed the highest training time, whereas different weight initiation techniques contribute in reducing the training times for the network. © 2024 The Authors.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

EFS-XGBoost: A robust framework for precision classification of COVID-19 cases

引用

Intelligent Data Analysis 2025年第1期29卷 171-185页

作者： Abidi, Mustufa Haider Khare, Neelu D, Preethi Alkhalefah, Hisham Umer, Usama Advanced Manufacturing Institute King Saud University Riyadh Saudi Arabia School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore India Department of Computer Science and Engineering Faculty of Engineering and Technology SRM Institute of Science and Technology Ramapuram Chennai India

The emergence of the novel COVID-19 virus has had a profound impact on global healthcare systems and economies, underscoring the imperative need for the development of precise and expeditious diagnostic tools. Machine learning techniques have emerged as a promising avenue for augmenting the capabilities of medical professionals in disease diagnosis and classification. In this research, the EFS-XGBoost classifier model, a robust approach for the classification of patients afflicted with COVID-19 is proposed. The key innovation in the proposed model lies in the Ensemble-based Feature Selection (EFS) strategy, which enables the judicious selection of relevant features from the expansive COVID-19 dataset. Subsequently, the power of the eXtreme Gradient Boosting (XGBoost) classifier to make precise distinctions among COVID-19-infected patients is *** EFS methodology amalgamates five distinctive feature selection techniques, encompassing correlation-based, chi-squared, information gain, symmetric uncertainty-based, and gain ratio approaches. To evaluate the effectiveness of the model, comprehensive experiments were conducted using a COVID-19 dataset procured from Kaggle, and the implementation was executed using Python programming. The performance of the proposed EFS-XGBoost model was gauged by employing well-established metrics that measure classification accuracy, including accuracy, precision, recall, and the F1-Score. Furthermore, an in-depth comparative analysis was conducted by considering the performance of the XGBoost classifier under various scenarios: employing all features within the dataset without any feature selection technique, and utilizing each feature selection technique in isolation. The meticulous evaluation reveals that the proposed EFS-XGBoost model excels in performance, achieving an astounding accuracy rate of 99.8%, surpassing the efficacy of other prevailing feature selection techniques. This research not only advances the field of COVI

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Autoencoder-Enhanced Roommate Recommendation System 6

Autoencoder-Enhanced Roommate Recommendation System

引用

6th IEEE Novel Intelligent and Leading Emerging sciences Conference, NILES 2024

作者： Zahran, Khaled Amir, Abdallah Elbanhawy, Magy Ghanem, Ingy El-Ghamry, Amir Fouad, Khaled M. Moawad, Ibrahim F. New Mansoura University Faculty of Computer Science and Engineering Egypt School of Engineering and Computer Science Mansoura University University of Hertfordshire Hosted by Global Academic Foundation Faculty of Computer and Information Sciences Egypt Benha University Information Systems Faculty of Computers and Artificial Intelligence Egypt College of Computer Science and Engineering Taibah University Department of Artificial Intelligence Saudi Arabia

ISBN: (纸本)9798350378511

This proposed system is designed for creating a new way of giving personalized recommendations by focusing on people's behaviors and preferences. The system uses traditional machine learning algorithms integrating it with deep learning techniques to get the most out of data and suggest recommendations which are designed according to each individual's unique preferences. A deep learning autoencoder is used to learn a lower-dimensional representation of the data, with an assurance on feature extraction and reconstruction accuracy. The encoded features are passed to be clustered using KMeans, with the effectiveness of clustering estimated through internal validation metrics such as silhouette score, Calinski-Harabasz index, and Davies-Bouldin index. Also, t-Distributed Stochastic Neighbor Embedding (t-SNE) is utilized for visualizing clustered data in a simple manner. Additionally, a silhouette plot is given to provide a visual representation of the silhouette scores across clusters, highlighting the degree of cohesion within clusters and separation between them. Finally, using cosine similarity that identifies similar users within the same cluster. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Stock Price Prediction: A Hybrid Approach Leveraging Large Language Models and Deep Learning

Enhancing Stock Price Prediction: A Hybrid Approach Leveragi...

引用

2024 International Conference on computer and Applications, ICCA 2024

作者： Ahmed, Mohamed Erfan Ebrahim, Gamal A. Abdelaal, Marwa A. Ain Shams University Faculty of Engineering Computer and Systems Engineering Department Cairo Egypt Sadat Academy for Management Sciences Faculty of Computers and Information Computer Science Department Cairo Egypt

ISBN: (纸本)9798350367560

Accurate stock price prediction is a challenging yet crucial goal in finance, with significant implications for investment decisions and risk management. This paper presents a comprehensive review of machine learning techniques for stock price prediction, examining traditional methods such as regression and ensemble models, as well as advanced approaches that integrate sentiment analysis and textual data sources. With the emergence of powerful Large Language Models (LLMs) such as ChatGPT, Llama and Gemini, we explore their potential for enhancing predictive accuracy using historical stock data. Key challenges are discussed, including data quality, model interpretability, and adapting to dynamic market conditions. Additionally, this paper proposes a trustworthy stock price prediction model based on LLMs enabling informed investment decision-making. Experimental results demonstrate that ChatGPT-4o model achieved a prediction accuracy of approximately 97%, which can be improved by tuning model parameters. Consequently, the paper highlights the potential of LLMs in improving stock price forecasting. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Identify the Intended Covert Value Among an Adversary in a (t, n) Threshold Setup 16

Identify the Intended Covert Value Among an Adversary in a (...

引用

16th IEEE International Conference on Computational Intelligence and Communication Networks, CICN 2024

作者： Biswas, Anindya Kumar Das, Tanisha Rana, Nilava Dasgupta, Mou Nit Raipur Computer Science and Business Systems Techno India University India Techno India University Computer Science and Engineering Kolkata India National Institute of Technology Chhattisgarh Raipur India

ISBN: (纸本)9798331505264

In safeguarding daily communications over open networks, ensuring data confidentiality is paramount to prevent sensitive information from reaching unintended recipients. Secret sharing techniques facilitate secure private key exchanges for data transmission over public networks. This paper introduces a two-phase protocol that allows honest participants to detect and counter malicious activities by reconstructing the intended covert value, even in presence of an unknown adversary who is one of the legitimate participants. A trusted Dealer, in a (t, n) threshold setup, generates and distributes shares using a randomly derived polynomial. Polynomial reconstruction is incorporated to help retrieve the original secret securely. The proposed methods employ hash-based mechanisms to ensure information security and integrity, with the second approach introducing additional randomness to enhance protection against insider attacks. Comprehensive security analysis underscores the protocol's resilience, demonstrating its effectiveness in maintaining confidentiality. © 2024 IEEE.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study on Deep Learning based Segmentation Techniques for Tamil Inscriptions 8

A Comparative Study on Deep Learning based Segmentation Tech...

引用

8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud), I-SMAC 2024

作者： Poornimathi, K. Muralibhaskaran, V. Priya, L. Rajalakshmi Engineering College Department of Computer Science and Business Systems India Rajalakshmi Engineering College Department of Computer Science and Engineering India Rajalakshmi Engineering College Department of Information Technology India

ISBN: (纸本)9798350376425

The digitization and preservation of Tamil inscriptions are crucial for safeguarding the rich cultural heritage they represent. This study presents an in-depth evaluation of deep learning-based segmentation methods specifically designed for the complex and ancient Tamil script. Evaluate several cutting-edge deep learning models, including Convolutional Neural Networks (CNNs), Fully Convolutional Networks (FCNs), U-Net, and attention-based mechanisms, using a dataset of Tamil inscriptions. The evaluation criteria encompass segmentation accuracy, computational efficiency, and the ability to handle the noise and degradation common in historical artifacts. Our findings indicate that traditional methods often falter when faced with the complexities of Tamil script, whereas advanced deep learning techniques markedly enhance segmentation quality. This improvement facilitates more precise transcription and analysis. The comparison emphasizes the benefits and challenges of each method, delivering important insights into their effectiveness for cultural heritage preservation and paving the way for future research in the digitization of ancient scripts. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Artificial Fish Swarm for Multi Protein Sequences Alignment in Bioinformatics

引用

computers, Materials & Continua 2022年第9期72卷 6091-6106页

作者： Medhat A.Tawfeek Saad Alanazi A.A.Abd El-Aziz Department of Computer Science College of Computer and Information SciencesJouf UniversitySaudi Arabia(KSA Department of Computer Science Faculty of Computers and InformationMenoufia UniversityEgypt Department of Information Systems College of Computer and Information SciencesJouf UniversitySaudi Arabia(KSA) Department of Information Systems and Technology Faculty of Graduates Studies and ResearchCairo UniversityEgypt

The alignment operation between many protein sequences or DNAsequences related to the scientific bioinformatics application is very *** is a trade-off in the objectives in the existing techniques of MultipleSequence Alignment (MSA). The techniques that concern with speed ignoreaccuracy, whereas techniques that concern with accuracy ignore speed. Theterm alignment means to get the similarity in different sequences with highaccuracy. The more growing number of sequences leads to a very complexand complicated problem. Because of the emergence;rapid development;anddependence on gene sequencing, sequence alignment has become importantin every biological relationship analysis process. Calculating the numberof similar amino acids is the primary method for proving that there is arelationship between two sequences. The time is a main issue in any alignmenttechnique. In this paper, a more effective MSA method for handling themassive multiple protein sequences alignment maintaining the highest accuracy with less time consumption is proposed. The proposed method dependson Artificial Fish Swarm (AFS) algorithm that can break down the mostchallenges of MSA problems. The AFS is exploited to obtain high accuracyin adequate time. ASF has been increasing popularly in various applicationssuch as artificial intelligence, computer vision, machine learning, and dataintensive application. It basically mimics the behavior of fish trying to getthe food in nature. The proposed mechanisms of AFS that is like preying,swarming, following, moving, and leaping help in increasing the accuracy andconcerning the speed by decreasing execution time. The sense organs that aidthe artificial fishes to collect information and vision from the environmenthelp in concerning the accuracy. These features of the proposed AFS make thealignment operation more efficient and are suitable especially for large-scaledata. The implementation and experimental results put the proposed AFS as afirst choice in th

关键词： Multiple sequence alignment swarm intelligence artificial fish swarm protein sequences

来源：评论

学校读者我要写书评

暂无评论

Optimization of Electrocardiogram Classification Using Dipper Throated Algorithm and Differential Evolution

引用

computers, Materials & Continua 2023年第2期74卷 2379-2395页

作者： Doaa Sami Khafaga El-Sayed M.El-kenawy Faten Khalid Karim Sameer Alshetewi Abdelhameed Ibrahim Abdelaziz A.Abdelhamid D.L.Elsheweikh Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Communications and Electronics Delta Higher Institute of Engineering and TechnologyMansoura35111Egypt Faculty of Artificial Intelligence Delta University for Science and TechnologyMansoura35712Egypt General Information Technology Department Ministry of DefenseThe Executive AffairsExcellence Services DirectorateRiyadh11564Saudi Arabia Computer Engineering and Control Systems Department Faculty of EngineeringMansoura UniversityMansoura35516Egypt Department of Computer Science Faculty of Computer and Information SciencesAin Shams UniversityCairo11566Egypt Department of Computer Science College of Computing and Information TechnologyShaqra University11961Saudi Arabia Department of Computer Science Faculty of Specific EducationMansoura UniversityEgypt

Electrocardiogram(ECG)signal is a measure of the heart’s electrical ***,ECG detection and classification have benefited from the use of computer-aided systems by *** goal of this paper is to improve the accuracy of ECG classification by combining the Dipper Throated Optimization(DTO)and Differential Evolution Algorithm(DEA)into a unified algorithm to optimize the hyperparameters of neural network(NN)for boosting the ECG classification *** addition,we proposed a new feature selection method for selecting the significant feature that can improve the overall *** prove the superiority of the proposed approach,several experimentswere conducted to compare the results achieved by the proposed approach and other competing ***,statistical analysis is performed to study the significance and stability of the proposed approach using Wilcoxon and ANOVA *** results confirmed the superiority and effectiveness of the proposed *** classification accuracy achieved by the proposed approach is(99.98%).

关键词： Electrocardiogram differential evolution algorithm dipper throated optimization neural networks

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：