检索结果-内蒙古大学图书馆

MuTCELM: An optimal multi-TextCNN-based ensemble learning for text classification

Heliyon 2024年第19期10卷 e38515页

作者： Agbesi, Victor Kwaku Chen, Wenyu Yussif, Sophyani Banaamwini Ukwuoma, Chiagoziem C. Gu, Yeong Hyeon Al-antari, Mugahed A. School of Computer Science and Engineering University of Electronic Science and Technology of China No. 2006 Xiyuan Ave West Hi-Tech Zone Sichuan Chengdu China Sichuan Engineering Technology Research Center for Industrial Internet Intelligent Monitoring and Application Chengdu University of Technology No. 610059 Sichuan Chengdu China Department of Artificial Intelligence and Data Science College of AI Convergence Daeyang AI Center Sejong University Seoul 05006 South Korea

Feature extraction plays a critical role in text classification, as it converts textual data into numerical representations suitable for machine learning models. A key challenge lies in effectively capturing both semantic and contextual information from text at various levels of granularity while avoiding overfitting. Prior methods have often demonstrated suboptimal performance, largely due to the limitations of the feature extraction techniques employed. To address these challenges, this study introduces Multi-TextCNN, an advanced feature extractor designed to capture essential textual information across multiple levels of granularity. Multi-TextCNN is integrated into a proposed classification model named MuTCELM, which aims to enhance text classification performance. The proposed MuTCELM leverages five distinct sub-classifiers, each designed to capture different linguistic features from the text data. These sub-classifiers are integrated into an ensemble framework, enhancing the overall model performance by combining their complementary strengths. Empirical results indicate that MuTCELM achieves average improvements across all datasets in accuracy, precision, recall, and F1-macro scores by 0.2584, 0.2546, 0.2668, and 0.2612, respectively, demonstrating significant performance gains over baseline models. These findings underscore the effectiveness of Multi-TextCNN in improving model performance relative to other ensemble methods. Further analysis reveals that the non-overlapping confidence intervals between MuTCELM and baseline models indicate statistically significant differences, suggesting that the observed performance improvements of MuTCELM are not attributable to random chance but are indeed statistically meaningful. This evidence indicates the robustness and superiority of MuTCELM across various languages and text classification tasks. © 2024 The Author(s)

关键词： Deep learning Ensemble learning Feature extraction Multi-TextCNN Text classification

来源：评论

学校读者我要写书评

暂无评论

Developing the Role of Firewalls in Enhancing Web Security for Wireless Networks

Developing the Role of Firewalls in Enhancing Web Security f...

引用

Smart Generation Computing, Communication and Networking (SMART GENCON), International Conference on

作者： Ritushree Narayan N. V. Balaji Kalanandhini G Parikshit N. Mahalle Javed Wasim Roma Singh Department of Coputing & Information Technology Usha Martin University Ranchi India Department of Computer Science Karpagam Academy of Higher Education Coimbatore Department of Electronics and Communication Engineering Prince Shri Venkateshwara Padmavathy Engineering College Chennai Department of Artificial Intelligence & Data Science Vishwakarma Institute of Information Technology Pune India Department of Computer Science Mangalayatan University Aligarh India Department of Computer Science Himalayan University Itanagar India

Firewalls play a critical role in controlling incoming and outgoing community site visitors and shielding a wireless network from malicious assaults. This technical summary focuses on the development of firewalls as a way to beautify the internet safety for Wi-Fi networks. Contemporary Wi-Fi networks are increasingly more vulnerable to malicious cyber-attacks. Firewalls are one of the simplest strategies of safety, and are able to restricting visitors among networks or from the net. They also can be used to come across and perceive malicious pastime. The development of firewalls has improved at a fast pace in current years. New functions along with Intrusion Detection and Prevention systems (IDS/IPS), URL filtering, virus scanning, user authentication, and application layer filtering, have vastly advanced the effectiveness of firewalls. Advanced strategies such as deploy Rector Reactive tracking (IRM), and community get right of entry to manage (NAC) also are getting used to detect suspicious community requests and put into effect coverage. NAC may be used to both control get admission to network resources, and respond to malicious site visitors. The enhanced internet protection supplied by firewalls is useful for both home and employer networks. Via implementing those capabilities, organizations can lessen the danger of cyberattacks and guard their resources.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Interactively Robot Action Planning with Uncertainty Analysis and Active Questioning by Large Language Model

arXiv

引用

arXiv 2023年

作者： Hori, Kazuki Suzuki, Kanata Ogata, Tetsuya Faculty of Science and Engineering Waseda University Tokyo169-8050 Japan Artificial Intelligence Laboratories Fujitsu Limited. Kanagawa211-8588 Japan Waseda University Tokyo169-8555 Japan National Institute of Advanced Industrial Science and Technology Tokyo100-8921 Japan

The application of the Large Language Model (LLM) to robot action planning has been actively studied. The instructions given to the LLM by natural language may include ambiguity and lack of information depending on the task context. It is possible to adjust the output of LLM by making the instruction input more detailed;however, the design cost is high. In this paper, we propose the interactive robot action planning method that allows the LLM to analyze and gather missing information by asking questions to humans. The method can minimize the design cost of generating precise robot instructions. We demonstrated the effectiveness of our method through concrete examples in cooking tasks. However, our experiments also revealed challenges in robot action planning with LLM, such as asking unimportant questions and assuming crucial information without asking. Shedding light on these issues provides valuable insights for future research on utilizing LLM for robotics. Copyright © 2023, The Authors. All rights reserved.

关键词： Uncertainty analysis

来源：评论

学校读者我要写书评

暂无评论

Automatic melanoma detection using discrete cosine transform features and metadata on dermoscopic images

引用

Journal of King Saud University - Computer and Information sciences 2024年第2期36卷

作者： Yousefi, Shamim Najjar-Ghabel, Samad Danehchin, Ramin Band, Shahab S. Hsu, Chung-Chian Mosavi, Amir Department of Computer Engineering University of Mohaghegh Ardabili Ardabil Iran Faculty of Electrical and Computer Engineering University of Tabriz Tabriz Iran International Graduate School of Artificial Intelligence National Yunlin University of Science and Technology Douliu Taiwan John von Neumann Faculty of Informatics Budapest Hungary Ludovika University of Public Service Budapest Hungary

Machine learning contributes in improving the accuracy of melanoma detection. There are extensive studies in classic and deep learning-based approaches for melanoma detection in the literature. Still, they are not accurate or require high learning data. This paper proposes a hybrid mechanism for automated melanoma detection on dermoscopic images based on Discrete Cosine Transform features and metadata. It is composed of three steps. First, extra information/artifacts are deleted;the remaining pixels are standardized for accurate processing. Second, the reliability of the mechanism is improved by the Radon transform, extra data is removed using the Top-hat filter, and the detection rate is increased by Discrete Wavelet Transform and Discrete Cosine Transform. Then, the number of features is reduced by Locality Sensitive Discriminant Analysis. The third step divides the images into learning and test ones to create image-based models using learning data. Finally, the best model is selected based on test data and improved by a metadata-based model. Simulation results show that the decision tree provides the most practical image-based model by improving accuracy and sensitivity. Besides, the comparison results demonstrate that our model improves the F-Value to detect melanoma superior to other approaches. © 2024 The Author(s)

关键词： artificial intelligence Big data data science Dermoscopic image Discrete cosine transform features Machine learning Melanoma detection Metadata

来源：评论

学校读者我要写书评

暂无评论

Neural Networks Based Text-Independent Speaker Identification with Breath and Ey

Neural Networks Based Text-Independent Speaker Identificatio...

引用

data science and Network Security (ICDSNS), International Conference on

作者： H S Ranjan Kumar Jyoti Metan Mahantesh Mathapati H P Bramha Prakash R J Kavitha D S Sunil Kumar Kiran Puttegowda Department of Artificial Intelligence & Data Science Shri Madhwa Vadiraja Institute of Technology and Management Udupi India Department of ISE Atria institute of Technology Bengaluru India Department of CSE Amruta Institute of Engineering and Management Sciences Bengaluru India Department of CSE Alva's Institute Of Engineering and Technology Moodbidri India Department of ECE East West Institute of Technology Bengaluru India Research Assistant Anatta innovations BGS B-School Bengaluru India Department of ECE Vidyavardhaka College of Engineering Mysuru India

ISBN: (数字)9798350373110

ISBN: (纸本)9798350373127

Identifying speakers in text-independent scenarios with short-time utterances poses challenges due to diverse acoustical variations, lack of contextual information, and limited data availability. Traditional speaker models like Gaussian Mixture Models rely on distribution assumptions and require substantial data for training, making them less suitable for scenarios with sparse or variable data. To address these challenges, this study proposes a neural network-based framework for speaker identification. The framework utilizes constant-Q spectrogram features extracted from speech signals and employs a combination of Convolutional Neural Networks (CNNs) and Long Short-Term Memory networks (LSTMs) for modeling. This approach aims to capture both spatial and temporal dependencies in speech features, thereby improving the robustness and accuracy of speaker identification. The effectiveness of the proposed framework is evaluated through experiments using small datasets containing breath or ‘Ey’ sound recordings. The results demonstrate that the framework achieves high accuracy in identifying speakers and exhibits rapid convergence during training. These findings highlight the potential of neural network-based approaches to overcome the limitations of traditional methods in speaker identification tasks with limited data. The neural network-based speaker identification framework presented in this study offers a promising alternative to Gaussian Mixture Models by leveraging deep learning techniques to handle diverse acoustical variations and small datasets effectively.

关键词： Training Adaptation models Accuracy Computational modeling Neural networks Feature extraction Robustness Convolutional neural networks Gaussian mixture model Spectrogram

来源：评论

学校读者我要写书评

暂无评论

Triply Laplacian Scale Mixture Modeling for Seismic data Noise Suppression

arXiv

引用

arXiv 2025年

作者： Pan, Sirui Zha, Zhiyuan Wang, Shigang Li, Yue Fan, Zipei Yan, Gang Nguyen, Binh T. Wen, Bihan Zhu, Ce College of Communication Engineering Jilin University Changchun130012 China School of Artificial Intelligence Jilin University Changchun130012 China College of Computer Science and Technology Jilin University Changchun130012 China Department of Computer Science Faculty of Mathematics and Computer Science University of Science Vietnam National University Ho Chi Minh City Ho Chi Minh City700000 Viet Nam School of Electrical and Electronic Engineering Nanyang Technological University 639798 Singapore Glasgow College University of Electronic Science and Technology of China Chengdu611731 China

Sparsity-based tensor recovery methods have shown great potential in suppressing seismic data noise. These methods exploit tensor sparsity measures capturing the low-dimensional structures inherent in seismic data tensors to remove noise by applying sparsity constraints through soft-thresholding or hard-thresholding operators. However, in these methods, considering that real seismic data are non-stationary and affected by noise, the variances of tensor coefficients are unknown and may be difficult to accurately estimate from the degraded seismic data, leading to undesirable noise suppression performance. In this paper, we propose a novel triply Laplacian scale mixture (TLSM) approach for seismic data noise suppression, which significantly improves the estimation accuracy of both the sparse tensor coefficients and hidden scalar parameters. To make the optimization problem manageable, an alternating direction method of multipliers (ADMM) algorithm is employed to solve the proposed TLSM-based seismic data noise suppression problem. Extensive experimental results on synthetic and field seismic data demonstrate that the proposed TLSM algorithm outperforms many state-of-the-art seismic data noise suppression methods in both quantitative and qualitative evaluations while providing exceptional computational efficiency. Copyright © 2025, The Authors. All rights reserved.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Can LVLMs Describe Videos like Humans? A Five-in-One Video Annotations Benchmark for Better Human-Machine Comparison

arXiv

引用

arXiv 2024年

作者： Hu, Shiyu Li, Xuchen Li, Xuzhao Zhang, Jing Wang, Yipei Zhao, Xin Cheong, Kang Hao School of Physical and Mathematical Sciences Nanyang Technological University Singapore Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China School of Automation Beijing Institute of Technology China Institute of Automation Southeast University Bangladesh School of Computer and Communication Engineering University of Science and Technology Beijing China College of Computing and Data Science Nanyang Technological University Singapore

Large vision-language models (LVLMs) have made significant strides in addressing complex video tasks, sparking researchers’ interest in their human-like multimodal understanding capabilities. Video description serves as a fundamental task for evaluating video comprehension, necessitating a deep understanding of spatial and temporal dynamics, which presents challenges for both humans and machines. Thus, investigating whether LVLMs can describe videos as comprehensively as humans—through reasonable human-machine comparisons using video captioning as a proxy task—will enhance our understanding and application of these models. However, current benchmarks for video comprehension have notable limitations, including short video durations, brief annotations, and reliance on a single annotator’s perspective. These factors hinder a comprehensive assessment of LVLMs’ ability to understand complex, lengthy videos and prevent the establishment of a robust human baseline that accurately reflects human video comprehension capabilities. To address these issues, we propose a novel benchmark, FIOVA (Five In One Video Annotations), designed to evaluate the differences between LVLMs and human understanding more comprehensively. FIOVA includes 3,002 long video sequences (averaging 33.6 seconds) that cover diverse scenarios with complex spatiotemporal relationships. Each video is annotated by five distinct annotators, capturing a wide range of perspectives and resulting in captions that are 4 ∼ 15 times longer than existing benchmarks, thereby establishing a robust baseline that represents human understanding comprehensively for the first time in video description tasks. Using the FIOVA benchmark, we conducted an in-depth evaluation of six state-of-the-art LVLMs (VideoLLaMA2, LLaVA-NEXT-Video, Video-LLaVA, VideoChat2, Tarsier, and ShareGPT4Video), comparing their performance with humans. Results show that while current LVLMs demonstrate some perception and reasoning capabilities, they st

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

The Smart Optimization and Estimating the Performance of Encapsulating Security Payload in Wireless Network Security

The Smart Optimization and Estimating the Performance of Enc...

引用

Smart Generation Computing, Communication and Networking (SMART GENCON), International Conference on

作者： Sripriya. T Monali R. Borade Neeta Deepaware Jambi Ratna Raja Kumar Md. Saifullah Khalid K. Ranjith Singh Department of Electronics and Communication Engineering Prince Shri Venkateshwara Padmavathy Engineering College Chennai Department of Artificial Intelligence & Data Science Vishwakarma Institute of Information Technology Pune India Department of Management Mangalayatan University Jabalpur India Department of Computer Engineering Genba Sopanrao Moze College of Engineering Balewadi Pune Maharashtra India Faculty of Business Mangagement & Commerce Usha Martin University Ranchi India Department of Computer Science Karpagam Academy of Higher Education Coimbatore

The Encapsulating Security Payload (ESP) is the primary issue of safety structure for wireless networks. It inspires to make specific cozy switches of records across a wireless network by offering encrypted messages, information integrity guarantee, and authentication of communications. ESP can also provide excellent selective filtering of community site visitors and provides levels of authentication required for ease of network access. Due to the advances in the wireless era, the overall performance of ESP in Wi-Fi networks is becoming increasingly essential. Consequently, intelligent optimization and estimating the overall performance of ESP in wireless networks is an important task. Clever optimization of ESP in Wi-Fi networks determines the most reliable parameters for ESP to maximize its performance and reduce the network overhead that may arise while looking to take care of records transmission and information protection. Traditional optimization techniques, including Genetic Algorithms, can be used for this project, as they give more accurate outcomes and might effortlessly be implemented in the community infrastructure. Clever optimization can't only decorate the performance of ESP but can also reduce the variety of false alarms in the machine. Moreover, developing an accurate estimation for the overall performance of ESP in Wi-Fi networks is essential to get the most out of its talents. The usage of mathematical models, including a Markov Chain modeling technique, to symbolize various.

关键词：

来源：评论

学校读者我要写书评

暂无评论

从信号到知识——基于人工智能的医学影像裸数据诊断价值初探

引用

engineering 2024年第3期34卷 60-69页

作者： Bingxi He Yu Guo Yongbei Zhu Lixia Tong Boyu Kong Kun Wang Caixia Sun Hailin Li Feng Huang Liwei Wu Meng Wang Fanyang Meng Le Dou Kai Sun Tong Tong Zhenyu Liu Ziqi Wei Wei Mu Shuo Wang Zhenchao Tang Shuaitong Zhang Jingwei Wei Lizhi Shao Mengjie Fang Juntao Li Shouping Zhu Lili Zhou Shuo Wang Di Dong Huimao Zhang Jie Tian Beijing Advanced Innovation Center for Big Data-Based Precision Medicine School of Engineering MedicineBeihang UniversityBeijing 100191China Key Laboratory of Big Data-Based Precision Medicine Beihang UniversityMinistry of Industry and Information TechnologyBeijing 100191China CAS Key Laboratory of Molecular Imaging Institute of AutomationChinese Academy of SciencesBeijing 100190China Department of Radiology The First Hospital of Jilin UniversityChangchun 130021China Neusoft Medical Systems Co.Ltd. Shenyang 110167China Neusoft Research of Intelligent Healthcare Technology Co.Ltd. Shenyang 110167China School of Life Science and Technology Xidian UniversityXi’an 710126China Engineering Research Center of Molecular and Neuro Imaging Ministry of EducationXi’an 710126China School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China

Encouraging and astonishing developments have recently been achieved in image-based diagnostic *** medical care and imaging technology are becoming increasingly ***,the current diagnosis pattern of signal to image to knowledge inevitably leads to information distortion and noise introduction in the procedure of image reconstruction(from signal to image).artificial intelligence(AI)technologies that can mine knowledge from vast amounts of data offer opportunities to disrupt established *** this prospective study,for the first time,we develop an AI-based signal-toknowledge diagnostic scheme for lung nodule classification directly from the computed tomography(CT)raw data(the signal).We find that the raw data achieves almost comparable performance with CT,indicating that it is possible to diagnose diseases without reconstructing ***,the incorporation of raw data through three common convolutional network structures greatly improves the performance of the CT models in all cohorts(with a gain ranging from 0.01 to 0.12),demonstrating that raw data contains diagnostic information that CT does not *** results break new ground and demonstrate the potential for direct signal-to-knowledge domain analysis.

关键词： Computed tomography Diagnosis Deep learning Lung cancer Raw data

来源：评论

学校读者我要写书评

暂无评论

Adversarial Attacks on Large Language Model-Based System and Mitigating Strategies: A Case Study on ChatGPT

引用

Security and Communication Networks 2023年第1期2023卷

作者： Liu, Bowen Xiao, Boao Jiang, Xutong Cen, Siyuan He, Xin Dou, Wanchun State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Guangdong Laboratory of Artificial Intelligence and Digital Economy Shenzhen China School of Computer Science & Technology Nanjing University of Posts and Telecommunications Nanjing China College of Big Data and Intelligent Engineering Southwest Forestry University Kunming China

Machine learning algorithms are at the forefront of the development of advanced information systems. The rapid progress in machine learning technology has enabled cutting-edge large language models (LLMs), represented by GPT-3 and ChatGPT, to perform a wide range of NLP tasks with a stunning performance. However, research on adversarial machine learning highlights the need for these intelligent systems to be more robust. Adversarial machine learning aims to evaluate attack and defense mechanisms to prevent the malicious exploitation of these systems. In the case of ChatGPT, adversarial induction prompt can cause the model to generate toxic texts that could pose serious security risks or propagate false information. To address this challenge, we first analyze the effectiveness of inducing attacks on ChatGPT. Then, two effective mitigating mechanisms are proposed. The first is a training-free prefix prompt mechanism to detect and prevent the generation of toxic texts. The second is a RoBERTa-based mechanism that identifies manipulative or misleading input text via external detection models. The availability of this method is demonstrated through experiments. © 2023 Bowen Liu et al.

关键词： Machine learning

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：