检索结果-内蒙古大学图书馆

3rd International Conference on Artificial Intelligence and Signal Processing, AISP 2023

作者： Duraisamy, Premkumar Natarajan, Yuvaraj Niranjani, V. Parvathy, K. KPR Institute of Engineering and Technology Department of Artificial Intelligence and Data Science Coimbatore India KPR Institute of Engineering and Technology Department of Computer Science and Engineering Coimbatore India Sri Eshwar College of Engineering Department of Computer Science and Engineering Coimbatore India Dr NGP Institute of Technology Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798350320749

Identifying gravitational waves produced by binary black hole mergers has sparked an unheard-of revolution in physics and astronomy. However, due to the low magnitudes of gravitational wave signals and the inevitability of detector noise, proper processing of gravitational wave data is in high demand. In this research, we propose a deep learning technique termed convolution neural network based on a set of time-series data from '***/datasets/andradaolteanu/g2net-gravitational-wave-dataset' that contains simulated gravitational waves with noise. This work uses a convolution neural network and Fourier transformation-based gravitational wave signal recognition model (CNN). According to experimental findings, the frequency-domain signal graph describes the gravitational wave signal's features more accurately than the time-domain signal. The signal recognition accuracy has been improved using CNN's convolutional processing. © 2023 IEEE.

关键词： Gravity waves

来源：评论

学校读者我要写书评

暂无评论

Scalable Algorithms for Densest Subgraph Discovery 39

Scalable Algorithms for Densest Subgraph Discovery

引用

39th IEEE International Conference on data engineering, ICDE 2023

作者： Luo, Wensheng Tang, Zhuo Fang, Yixiang Ma, Chenhao Zhou, Xu Hunan University College of Computer Science and Electronic Engineering China The Chinese University of HongKong School of Data Science Shenzhen China

ISBN: (纸本)9798350322279

As a fundamental problem in graph data mining, Densest Subgraph Discovery (DSD) aims to find the subgraph with the highest density from a graph. It has been studied for several decades and found a large number of real-world applications, such as network community detection, regulatory motif discovery in DNA, graph index construction, and fake follower detection. Although there are many existing DSD algorithms, they are often not scalable or efficient to process large-scale graphs, since most of them are serial algorithms and can only leverage the computing resource of a single CPU core. To tackle these issues, in this paper we propose efficient parallel algorithms for solving the DSD problems on both undirected and directed graphs at scale. Our main idea is to use the k-cores (a kind of dense subgraph) to approximate the densest subgraph in the undirected graphs, and then propose efficient parallel algorithms for computing the cores by optimizing the iterative process and also reducing the number of iterations. We further extend this idea for directed graphs by introducing a novel concept, named w-induced subgraph, to avoid unnecessary enumerations of x or y when searching [x,y]-cores (a kind of directed dense subgraph to approximate the densest). To verify the scalability and efficiency of the proposed algorithms, we have conducted extensive experiments on 12 large real-world graphs, and four of them are billion-scale. The experimental results show that our proposed algorithms outperform the state-of-the-art algorithms on both undirected and directed graphs, in terms of scalability and efficiency. © 2023 IEEE.

关键词： Directed graphs

来源：评论

学校读者我要写书评

暂无评论

Multi-View Multi-Modal Head-Gaze Estimation for Advanced Indoor User Interaction

引用

computers, Materials & Continua 2022年第3期70卷 5107-5132页

作者： Jung-Hwa Kim Jin-Woo Jeong Department of Computer Engineering Kumoh National Institute of TechnologyGumi39177Korea Department of Data Science Seoul National University of Science and TechnologySeoul01811Kore

Gaze estimation is one of the most promising technologies for supporting indoor monitoring and interaction ***,previous gaze estimation techniques generally work only in a controlled laboratory environment because they require a number of high-resolution eye *** makes them unsuitable for welfare and healthcare facilities with the following challenging characteristics:1)users’continuous movements,2)various lighting conditions,and 3)a limited amount of available *** address these issues,we introduce a multi-view multi-modal head-gaze estimation system that translates the user’s head orientation into the gaze *** proposed system captures the user using multiple cameras with depth and infrared modalities to train more robust gaze estimators under the aforementioned *** this end,we implemented a deep learning pipeline that can handle different types and combinations of *** proposed system was evaluated using the data collected from 10 volunteer participants to analyze how the use of single/multiple cameras and modalities affect the performance of head-gaze *** various experiments,we found that 1)an infrared-modality provides more useful features than a depth-modality,2)multi-view multi-modal approaches provide better accuracy than singleview single-modal approaches,and 3)the proposed estimators achieve a high inference efficiency that can be used in real-time applications.

关键词： Human-computer interaction deep learning head-gaze estimation indoor monitoring

来源：评论

学校读者我要写书评

暂无评论

Classification of hyperspectral images using fusion of CNN and MiniGCN with SVM

引用

International Journal of Digital Earth 2023年第1期16卷 3601-3617页

作者： Wenbing Wu Tariq Sadad Mejdl Safran Sultan Alfarhood Xiaojian Yuan Big Data College Fuzhou University of International Studies and TradeFuZhouPeople’s Republic of China Department of Computer Science University of Engineering&TechnologyMardanPakistan Department of Computer Science College of Computer and Information SciencesKing Saud UniversityP.O.Box 51178Riyadh11543Saudi Arabia

Convolutional neural networks(CNNs)have gained popularity for categorizing hyperspectral(HS)images due to their ability to capture representations of spatial-spectral ***,their ability to model relationships between data is *** convolutional networks(GCNs)have been introduced as an alternative,as they are effective in representing and analyzing irregular data beyond grid *** have *** computationally intensive,minibatch GCNs(miniGCNs)enable minibatch training of large-scale *** have improved the classification performance by using miniGCNs to infer out-of-sample data without retraining the *** addition,fuzing the capabilities of CNNs and GCNs,through concatenative fusion has been shown to improve performance compared to using CNNs or GCNs ***,support vector machine(SvM)is employed instead of softmax in the classification *** techniques were tested on two HS datasets and achieved an average accuracy of 92.80 using Indian Pines dataset,demonstrating the effectiveness of miniGCNs and fusion strategies.

关键词： CNN hyperspectral images MiniGCN SVM

来源：评论

学校读者我要写书评

暂无评论

Predicting Employee Attrition: A Comparative Analysis of Machine Learning Models Using the IBM Human Resource Analytics dataset

引用

Procedia computer science 2025年 258卷 4084-4093页

作者： Rajkumar Govindarajan N. Komal Kumar Sudhakar Reddy P Sai Pravallika E Dhatri B Pavan Kumar G Department of Computer Science & Engineering (Data Science) Madanapalle Institute of Technology & Science Madanapalle Andhra Pradesh-517

In employee turnover research and workforce management, addressing the impacts of suboptimal employee performance is crucial for organizations of all sizes and industries. Utilizing advanced machine learning classification models to predict potential employee resignations can enhance human resource departments’ intervention strategies, effectively mitigating attrition challenges. This research investigates the performance of various machine learning algorithms in classification tasks, focusing on their accuracy in predicting outcomes from a given dataset. Five models were evaluated: Random Forest, Support Vector Machine (SVM), Decision Tree, Gradient Boosting, and a Hybrid Model that integrates multiple algorithms. The goal was to identify which model yields the highest accuracy and to understand the strengths and weaknesses of each approach. Results showed that the Hybrid Model achieved the highest accuracy at 95.0%, suggesting that combining different algorithms effectively harnesses their strengths while mitigating individual weaknesses. The SVM accurately classified the instance with 88.6%, demonstrating its capability to manage complex decision boundaries in high-dimensional spaces. Both Random Forest and Gradient Boosting attained an accuracy of 87.3%, reflecting their ensemble techniques that enhance predictive performance by reducing overfitting and optimizing error reduction. In comparison, the Decision Tree classifier exhibited the least accuracy at 80.5%, highlighting its susceptibility to overfitting and limited generalizability. The superior performance of the Hybrid Model indicates a promising direction for future research, where integrating diverse algorithms could lead to more robust predictions. Overall, this study provides valuable insights for practitioners and researchers seeking to optimize model selection and improve predictive accuracy in their domains.

关键词： Employee turnover Workforce Predictive Human Resourse Intervention

来源：评论

学校读者我要写书评

暂无评论

Analysis of MFCC and Chromagram Features in Lightweight Bidirectional LSTM and Attention Models for Music Genre Classification 21

Analysis of MFCC and Chromagram Features in Lightweight Bidi...

引用

21st International computer Conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2024

作者： Huang, Renxiang Li, Xu Li, Jianping School of Artificial Intelligence and Big Data Sichuan University of Arts and Science China Khoury College of Computer Science Northeastern university BostonMA02115 United States School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu610054 China

ISBN: (纸本)9798331519254

Music genre classification is essential for organizing music libraries and enhancing recommendation systems. This paper evaluates four lightweight models combining Mel Frequency Cepstral Coefficients (MFCCs) and Chromagram features with Bidirectional LSTM and Multi-head Attention architectures. Using the GTZAN dataset, we found that models with Multi-head Attention, outperformed with MFCCs features, achieving a test accuracy of 85.6%. MFCCs excel in capturing timbral characteristics critical for genre classification, while Chromagram features proved less effective. Although overfitting occurred, future studies can focus on incorporating methods like data augmentation and regularization to improve performance. These lightweight models offer a balance between accuracy and computational efficiency, making them suitable for real-time applications. © 2024 IEEE.

关键词： Real time systems

来源：评论

学校读者我要写书评

暂无评论

Effective Early Hypothyroid Disease Prediction Using Traditional and Ensemble Machine Learning Algorithms 15th

Effective Early Hypothyroid Disease Prediction Using Traditi...

引用

15th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2023

作者： Mridul, Aunik Hasan Ahsan, Nowreen Saleh, Md. Abu Asif, Asifuzzaman Afrose, Sonia Computer Science and Engineering Daffodil International University Dhaka Bangladesh Department of Data Science Clarkson University New York United States Computer Science and Engineering Lovely Professional University Punjab India Nutrition and Food Engineering Daffodil International University Dhaka Bangladesh

ISBN: (纸本)9783031810794

In recent years, there has been a noticeable rise in the prevalence of physical ailments, most notably hypothyroidism, a condition that has garnered substantial attention due to its substantial impact on a significant portion of the population. To gain a comprehensive understanding of the disease’s severity, distinguishing between standard and affected diagnostic reports is imperative. In this work, we suggest using algorithmic models to facilitate early identification and increase knowledge of possible health hazards linked to hypothyroidism. Our approach is straightforward and well-suited for the prediction of uncomplicated cases of hypothyroid illness in real-world scenarios. Our dataset, sourced from Kaggle, served as the foundation for our research, which involved the utilization of a variety of classifiers, including RF, LR, GB, KN, ABC, DT, GS, and ensemble techniques. The DT classifier proved to be the most accurate, with an astounding accuracy rate of 98.278%. The findings were quite successful. In addition, the accuracy of the Voting Classifier RDGL was 97.218%, while the Stacking Classifier RDAS demonstrated an accuracy of 97.483%. Our optimization efforts, which included hyperparameter tuning, further enhanced the performance of each classifier. Based on extensive experimentation and a review of contemporary research, our findings unequivocally endorse the Decision Tree (DT) boosting classifier as exceptionally proficient, demonstrating a remarkable accuracy rate of 98.278% in the precise prediction of hypothyroid disease. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Bagging Boosting Ensemble Hypothyroidism

来源：评论

学校读者我要写书评

暂无评论

CNN-DBO: Dung Beetle Optimization Based Urban Area Change Detection Using Satellite Images

CNN-DBO: Dung Beetle Optimization Based Urban Area Change De...

引用

2025 IEEE International Conference on Computational, Communication and Information Technology, ICCCIT 2025

作者： Pujari, Jambukeshwar Wasim, Javed Tripathi, Aprna Mangalayatan University Department of Computer Engineering and Applications Aligarh20221385 India Mangalayatan University Department of Computer Engineering and Applications Aligarh India Manipal University Department of Data Science Engineering Jaipur India

ISBN: (纸本)9798331512965

In recent years, Geographic Information Systems (GIS) have garnered a significant deal of interest for their ability to detect changes in metropolitan areas. One of the uses of change detection in satellite photographs is in the field of urban planning. Other applications include environmental monitoring and remote surveillance. A number of different approaches have been suggested in recent times for the purpose of automatically identifying alterations in the land cover based on GIS photographs. However, due to poor texture, cluttered context, and considerable object scale disparities, the convolutional neural network (CNN) method is only partially successful in capturing the context from geographic information systems (GISs). This is despite the fact that it remains an effective method for detecting change in urban environments. The article presents the construction of a CNN-based change detection model using the utilization of Dung Beetle Optimization (DBO). The DBO algorithm is used to adjust the loss parameter in order to increase the overall performance. The approach that was proposed makes use of the Onera satellite data set and displays an overall accuracy of 97.50% percent. In light of this, it appears that the strategy that is proposed is both very efficient and effective. © 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

Identifying player skill of dota 2 using machine learning pipeline

引用

Discover Artificial Intelligence 2024年第1期4卷 41页

作者： Pengmatchaya, Methasit Natwichai, Juggapong Data Science Consortium Chiang Mai University Huay Kaew Muang Chiang Mai50200 Thailand Department of Computer Engineering Faculty of Engineering Chiang Mai University Huay Kaew Muang Chiang Mai50200 Thailand

The esports industry is one of the prominent business sectors in the digital era, particularly, Multiplayer Online Battle Arena (MOBA) games which gain much attention from gamers and streaming audiences. Among such games, Defense of the Ancient 2 or Dota 2 is the record holder for the highest prize esports tournament. Therefore, various companies and investors start their esports teams to compete in the Dota 2 tournaments, the Internationals. To success in the competition, player recruitment is a crucial process as it usually takes considerable effort to find a skillful player. Watching the game replay to evaluate the player’s skill is one of the approaches. However, it can be too exhaustive, also some player’s ranking, which represent the player’s skill, are often not available. In this paper, we propose an effective machine learning pipeline to evaluate the player’s skill. Our designed pipeline includes data collection, feature engineering, and machine learning modeling. We show the data collection process using open-source API. An effective method for feature engineering is proposed. Features, e.g., end-game, or tactical decision related statistics, are incorporated along with the resource in the game distribution, harassment tactic, or spatiotemporal features, in order to provide effective models. Subsequently, we apply major machine learning models based on a single game data, i.e., logistic regression and random forest, to the processed data. The most effective model can achieve up to 0.7091 precision, 0.5850 recall, 0.6411 F1-score, and 0.8123 ROC AUC score. © The Author(s) 2024.

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

Revealing Security Flaws in Cross-Modal Retrieval Models through Video Poisoning

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第6期35卷 6184-6194页

作者： Jin, Ming Hu, Wenbo Hong, Richang Zhu, Lei Hefei University of Technology School of Computer and Information Hefei China Data Space Research Institute Hefei Comprehensive National Science Center China Tongji University School of Electronic and Information Engineering Shanghai China

Video-text cross-modal retrieval is widely studied to improve retrieval accuracy. However, the security of video-text cross-modal retrieval models receives little attention. If attackers exploit the security vulnerabilities in these models, it poses a significant threat to the retrieval models. Thus, identifying security flaws in video-text cross-modal retrieval models becomes the focus of our research. We are the first to design a video poisoning model to uncover security vulnerabilities in retrieval models. Existing poisoning models have certain limitations when it comes to exploiting vulnerabilities in retrieval models. These include failing to comprehensively embed malicious information into the original video and being unable to maintain visual consistency between the original and poisoned videos. These limitations can result in unsuccessful attacks on retrieval models and an inability to effectively identify security flaws within them. To address these shortcomings, we design an efficient poisoning model that embeds malicious information thoroughly into the original clean data to attack video-text cross-modal retrieval models. We are the first to use a poisoning model to attack retrieval models, thereby uncovering their security vulnerabilities. Second, we introduce a bi-level poisoning module to ensure that malicious information is thoroughly embedded into the original video, thereby enhancing the attack capability of the poisoning model. Finally, we design an adversarial module to improve visual consistency between the original and poisoned videos, thus enhancing the concealment of malicious information within the training data of retrieval models. Our poisoning model can identify security flaws in video-text cross-modal retrieval models, providing insights into improving the security of retrieval models. The effectiveness of our model is validated on the MSR-VTT, LSMDC, and MSVD datasets. © 1991-2012 IEEE.

关键词： Content based retrieval

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：