检索结果-内蒙古大学图书馆

4th international conference on Sentiment Analysis and Deep learning, ICSADL 2025

作者： Divya, S. Sree Sastha Institute of Engineering and Technology Department of Computer Science and Engineering Tamil Nadu Chembarambakkam600123 India

ISBN: (数字)9798331523923

ISBN: (纸本)9798331523923

these days, malware evolves and multiplies exponentially through structural changes and camouflage using methods like encryption, obfuscation, polymorphism, and metamorphism. As deep learning has advanced, techniques like convolutional neural networks (CNN) have become powerful instruments for identifying complex patterns in this malicious software. the present study leverages CNN's capacity to detect patterns in malware datasets generated from RGB or images in greyscale and to determine the global structure of code that has been converted into an image. Convolutional Neural Networks (CNN) are a method of deep learning that has recently demonstrated better performance than conventional learning algorithms, particularly in applications like image categorization. Motivated by this result, a CNN-based malware sample categorisation architecture is proposed. After converting binaries of malware to monochrome images, we train a CNN to classify the images. Our method exceeds the latest advances, according to analyses done on the difficult malware classification dataset Malimg, © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Leveraging LLM-Powered Chatbot for learning and optimization in Databases and Information Systems Education 4

Leveraging LLM-Powered Chatbot for Learning and Optimization...

引用

4th international conference on Sentiment Analysis and Deep learning, ICSADL 2025

作者： Nageshwaran, Vinoth Nageshwaran, Sathish Business Insider New York United States Atos

ISBN: (纸本)9798331523923

this research work proposes a LLM-powered chatbot aimed at supporting students and educators in databases and information systems in higher education. the chatbot leverages the capabilities of LLM to facilitate interactive learning, offering real-time assistance on various database-related topics, including query formulation, data modeling, and system design. the chatbot incorporates advanced pre-processing techniques such as text cleaning, tokenization, stop words removal, and stemming/lemmatization to prepare user inputs for effective analysis. Feature extraction methods such as Bag-of-Words (BOW), n-grams, and Improved Term Frequency-Inverse Document Frequency (ITF-IDF) convert textual data into numerical forms, enhancing model understanding and accuracy. Additionally, deep learning algorithms, Hybrid Convolutional Long Short-Term Memory (HCLSTM) networks, are integrated to enhance the chatbot's ability to generate contextually relevant responses. optimization is achieved through the Improved Artificial Bee Colony (IABC) algorithm, which fine-tunes the chatbot's parameters for superior performance. through an interactive user interface, the chatbot provides personalized learning experiences, helping students grasp complex concepts in databases and information systems. Initial testing indicates the system's effectiveness in delivering insightful, tailored responses, making it a valuable tool for enhancing the learning process in higher education. © 2025 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Inter-layer Recurrent Neural Networks learning algorithms 4

Inter-layer Recurrent Neural Networks Learning Algorithms

引用

4th international conference on Computer, Communication, Control and Information Technology, C3IT 2024

作者： Ganguly, Priyam Ghosh, Akhilbaran Hanwha Q Cells America Inc. IrvineCA92618 United States Tata Consultancy Services Limited Mumbai India

ISBN: (纸本)9798350389128

In recent years, deep learning has shown remarkable progress in applications such as image recognition, speech recognition, games, and photorealistic image generation. this paper focuses on three core aspects of the study of artificial neural networks: expressivity, learnability, and interpretability. We introduce a novel architecture for deep recurrent neural networks called inter-layer recurrent neural networks (ILRNN). these networks exhibit extended and top-down connectivity across time steps without introducing cyclical paths in the computational graph. We demonstrate that recurrent neural networks can represent any Boolean circuit, interpreting ANNs as trainable versions of combinational logic. Using a binary addition task as a test bed, We show that while ILRNNs can theoretically express exact solutions, conventional training algorithms struggle to find the optimal weight configuration. our findings highlight the challenges in the learnability of such networks and suggest directions for future research, including alternative learning algorithms and integration with other deep learning advancements. © 2024 IEEE.

关键词： Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Hyperparameter optimization in CNN: A Review 4

Hyperparameter Optimization in CNN: A Review

引用

4th IEEE international conference on Computing, Communication, and Intelligent Systems, ICCCIS 2023

作者： Sharma, Anu Kumar, Dharmender Guru Jambheshwar University of Science and Technology Department of CSE Haryana Hisar India

ISBN: (纸本)9798350306118

Hyperparameter optimization is an important issue in convolutional neural networks (CNNs), which is an appropriate deep learning network for image classification. Several classical and metaheuristic algorithms are often employed to optimize hyperparameters. In the present paper, different algorithms are compared by using various evaluation measures. the literature reveals that genetic algorithms (GA) and PSO or particle swarm optimization are the most effective algorithms for hyperparameter optimization used in a variety of domains including disease diagnosis;face detection, and handwritten character recognition. Further, it is concluded that optimized CNNs, are essential to the early detection of a variety of diseases, which can assist physicians and clinicians in saving lives by providing accurate and timely predictions. © 2023 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Research on Energy Efficiency optimization of IoT Nodes Based on Event-Triggered Pulse Control 4

Research on Energy Efficiency Optimization of IoT Nodes Base...

引用

4th international conference on Communication Technology and Information Technology, ICCTIT 2024

作者： Chen, Zhongneng Westa College Southwest University Chongqing China

ISBN: (纸本)9798331528973

the advent of the Internet of things (IoT) has brought with it a rapid proliferation of applications, which has in turn created a significant challenge for IoT nodes operating in resource-constrained environments in terms of energy efficiency. In this paper, we put forward a model for optimising the energy efficiency of IoT nodes based on event-triggered pulse control. Based on the existing pulse-width modulation (PWM) and eventtriggered control models, we propose a novel hybrid method that combines event-triggered control and adaptive pulse modulation to dynamically adjust the energy consumption of node activities. the model employs a multi-level control architecture that integrates real-time monitoring, threshold-based event detection, and machine learning algorithms for adaptive control parameter tuning, thereby optimising energy utilisation and computational load. the experimental results demonstrate that the proposed model exhibits a 30% improvement in energy efficiency compared to existing energy efficiency optimisation models, while maintaining the same level of performance and reliability. © 2024 IEEE.

关键词： Pulse width modulation

来源：评论

学校读者我要写书评

暂无评论

Surface Defect Detection Based on Label Reassembling 4

Surface Defect Detection Based on Label Reassembling

引用

4th IEEE international conference on Power, Electronics and Computer applications, ICPECA 2024

作者： Liu, Gaokai Huzhou Institute of Zhejiang University Intelligent Sensing Laboratory Huzhou China

ISBN: (纸本)9798350359589

Owing to the complexity of surface defect morphologies, deep learning framework-based algorithms are usually able to obtain higher accuracy compared to traditional methods with hand-crafted features and gradually attract researchers' attention. However, considering the existing defect image data sets are usually small samples, overfitting is particularly liable to occur when the general classification model is adopted directly. To tackle this problem, we present a label reassembling-based surface defect detection. Specifically, the original label is reassigned to constructing a multi-task learning model, where the recombined label corresponds to the upstream auxiliary task to alleviate the overfitting problem. Finally, the experimental results on public datasets indicate that the improved method can reach better performance compared with MobileNetV2 which is used as the base model in this paper. © 2024 IEEE.

关键词： deep learning label reassembling surface defect detection

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning in wind energy - a review

引用

international JOURNAL OF GREEN ENERGY 2024年第9期21卷 1945-1968页

作者： Narayanan, Valayapathy Lakshmi Indian Inst Technol Guwahati Ctr Intelligent Cyber Phys Syst Res Bldg 4th floor Gauhati 781039 Assam India

Today's environmental concerns, particularly those related to global warming, have sparked a drive for the usage of renewable energy sources. One of the most significant sources of renewable energy is wind energy and wind energy conversion system are preferred to harvest wind energy. Due to the growing sophistication of wind energy conversion systems, new strategies based on advanced analytics are needed. In this study, reinforcement learning implemented in wind energy has been reviewed, the most popular approaches for various applications are identified, and it has been shown that reinforcement learning may be utilized in place of traditional approaches. According to the application, the techniques are examined and divided into four groups: optimal control, prediction and forecasting, optimization, and other techniques. Consequently, many literature has reported that, on an average, reinforcement learning has improved performance by 5% to 20% than existing methods. Moreover, around 85% of the 153 references included in this article were published after 2018. the purpose of the work is to provide a basis for future research on reinforcement learning applied in wind energy that may be crucial to energy sustainability. the report also addresses the discussion on the reinforcement learning current state, limitations, and future scope.

关键词： Reinforcement learning wind energy conversion system wind turbine optimization control strategies

来源：评论

学校读者我要写书评

暂无评论

Optimized algorithms for Zigzag and Contour Parallel Hatching in Industrial applications 4

Optimized Algorithms for Zigzag and Contour Parallel Hatchin...

引用

4th international conference on Ubiquitous Computing and Intelligent Information Systems, ICUIS 2024

作者： Vangala, Tejam Reddy Cleveland State University OH United States

ISBN: (纸本)9798331529635

the paper presents the development and implementation of advanced algorithms for zigzag and contour parallel hatching techniques used in industrial applications such as laser marking. the primary objective is to enhance the efficiency and accuracy of filling two-dimensional closed curve polygons. the proposed methodology includes two algorithms: a zigzag hatch approach that processes point data through vector transformations and spacing adjustments and a contour parallel hatch that generates points inward to refine edge accuracy. Both algorithms were validated in experimental setups focused on reducing idle time and improving precision in path generation. the proposed algorithms address key challenges such as reducing the idle time of robotic arms, minimizing data points, and optimizing multi-process operations. through a series of experimental validations, the algorithms demonstrate significant improvements in operational performance and resource utilization. the study employs two novel algorithms: the zigzag hatch and the contour parallel hatch, which process point data through vector transformations and spacing adjustments. these methods were validated through experimental setups aimed at evaluating accuracy and efficiency improvements in robotic laser applications. © 2024 IEEE.

关键词： Contour Parallel Hatch Industrial Laser Marking Path optimization Polygon Filling Technique Robotic Arm Efficiency Zigzag Hatch Algorithm

来源：评论

学校读者我要写书评

暂无评论

Predicting Retail Store Transaction Patterns: A Comparison of ARIMA and Machine learning Models 1

引用

4th international conference on optimization learning algorithms and applications

作者： Vaz, Clara B. Sena, Ines Braga, Ana Cristina Novais, Paulo Fernandes, Florbela P. Lima, Jose Pereira, Ana I. Inst Politecn Braganca Res Ctr Digitalizat & Intelligent Robot CeDRI Lab Assoc Sustentabilidade & Tecnol Regioes Monta Campus Santa Apolonia P-5300253 Braganca Portugal Univ Minho ALGORITMI Res Ctr LASI Campus Gualtar P-4710057 Braga Portugal

ISBN: (数字)9783031774263

ISBN: (纸本)9783031774256;9783031774263

Retail transactions represent sales of consumer goods, or final goods, by consumer companies. this sector faces security challenges due to the hustle and bustle of sales, affecting employees' workload. In this context, it is essential to estimate the number of customers who will appear in the store daily so that companies can dynamically adjust employee schedules, aligning workforce capacity with expected demand. this can be achieved by forecasting transactions using past observations and forecasting algorithms. this study aims to compare the ARIMA time series algorithm with several Machine learning algorithms to predict the number of daily transactions in different store patterns, considering data variability. the study identifies four typical store patterns based on these criteria using daily transaction data between 2019 and 2023 from all retail stores of the leading company in Portugal. Due to data variability and the results obtained, the algorithm that presents the most minor errors in predicting daily transactions is selected for each store. this study's ultimate goal is to fill the gap in forecasting daily customer transactions and present a suitable forecasting model to mitigate risks associated with transactions in retail stores.

关键词： Forecasting Transactions Time series Machine learning

来源：评论

学校读者我要写书评

暂无评论

An Improved Slime Mould optimization Algorithm Based on Integrating Multiple Strategies and Its applications 4

An Improved Slime Mould Optimization Algorithm Based on Inte...

引用

4th IEEE international conference on Data Science and Computer Application, ICDSCA 2024

作者： Tan, Weixiong Ling, Ping College of Intelligent Education Jiangsu Normal University Xuzhou China

ISBN: (纸本)9798350368239

In this paper, a multi-strategy enhanced slime mould algorithm (IMSMA) is proposed to address issues such as population initialization, convergence speed, and the tendency to fall into local optima in the slime mould optimization algorithm (SMA). Firstly, an optimal point set is introduced for population initialization to enhance initial diversity and improve algorithm stability. Secondly, the fixed parameter controlling population randomness is adjusted to adaptively change with the iteration number, balancing the exploration capability at the beginning phases and the optimization capability in the final phases. Additionally, a convergence stagnation judgment mechanism is incorporated to prevent premature convergence. Comparative experiments with similar intelligent optimization algorithms and the original SMA on functions from CEC2014 demonstrate that the enhanced algorithm achieves better stability and accuracy. Finally, the improved algorithm is applied to search for SVM kernel parameters, and experiments on eight UCI datasets further validate the feasibility and practicality of the enhanced algorithm. ©2024 IEEE.

关键词： optimization algorithms

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：