检索结果-内蒙古大学图书馆

Nature Inspired Optimization in Context-Aware-Based Coronary Artery Disease Prediction: A Novel Hybrid Harris Hawks Approach

引用

IEEE ACCESS 2024年 12卷 92635-92651页

作者： Ragavi Vijayaraj, Anu Pasupathi, Subbulakshmi Vellore Inst Technol Sch Comp Sci & Engn Chennai 600127 India

Coronary Artery Disease (CAD) imposes a significant global health burden, profoundly impacting morbidity and mortality rates worldwide. Accurate prediction of CAD is paramount for efficient management and prevention of associated complications. This study introduces a novel Hybrid Harris Hawks Optimization (H-HHO) approach, incorporating three noteworthy enhancements to augment classifier efficacy in CAD prediction compared to the conventional HHO algorithm. The advanced methodology was deployed for hyperparameter tuning of standard classification algorithms, including Support Vector Machine (SVM), Logistic Regression (LR), K-Nearest Neighbors (KNN), and Random Forest (RF). Moreover, a Context-Aware based Model (CAM) was employed to discern critical features (e.g., thallium and chest pain type) for CAD prediction, with subsequent comparison of their outcomes. The UCI heart disease dataset served as the basis for evaluating the efficiency of HHO and H-HHO algorithms, where H-HHO demonstrated superior performance, achieving an accuracy of 94.74% with LR and SVM, compared to the highest accuracy of 82.46% among classifiers using the HHO approach. The proposed H-HHO methodology for hyperparameter tuning in machine learning algorithms presents a promising framework, showcasing its effectiveness in CAD prediction. Future research endeavors may further explore H-HHO's application across diverse medical prediction tasks and its integration into other meta-heuristic algorithms to advance healthcare applications.

关键词： Diseases Optimization Accuracy Solid modeling Prediction algorithms classification algorithms CADCAM Metaheuristics Coronary arteriosclerosis CAD CAM classifiers meta-heuristics prediction

来源：评论

学校读者我要写书评

暂无评论

Automatically Evolving Interpretable Feature Vectors Using Genetic Programming for an Ensemble Classifier in Skin Cancer Detection

引用

IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE 2024年第3期19卷 26-41页

作者： Ul Ain, Qurrat Al-Sahaf, Harith Xue, Bing Zhang, Mengjie Victoria Univ Wellington Wellington New Zealand AlShaab Univ Baghdad Iraq

Early skin cancer diagnosis saves lives as the disease can be successfully treated through complete excision. Computer-aided diagnosis methods are developed using artificial intelligence techniques to help earlier detection and identify hidden causes leading to cancers in skin lesion images. In skin cancer image classification problems, an ensemble of classifiers has demonstrated better classification ability than a single classification algorithm. Traditionally, training an ensemble uses the complete set of original features, where some of these features can be redundant or irrelevant and hence, may not provide useful information in generating good models for ensemble classification. Moreover, newly created features may help improve classification performance. To address this issue, the existing methods have used feature construction for building an ensemble classifier, which usually creates a fixed number of features that may fit the training data too well, resulting in poor test performance. This study develops a novel classification approach that combines ensemble learning, feature selection, and feature construction utilizing genetic programming (GP) to handle the above limitations. The proposed method automatically evolves variable-length feature vectors consisting of GP-selected and GP-constructed features suitable for training an ensemble classifier. This study evaluates the effectiveness of the proposed method on two benchmark real-world skin image datasets that include dermoscopy and standard camera images. The experimental results reveal that the proposed algorithm significantly outperforms four state-of-the-art convolutional neural network methods, the existing GP approaches, and 11 commonly used machine learning methods. Furthermore, this study also includes interpreting evolved individuals that highlight important skin cancer characteristics playing a vital role in discriminating images of different cancer classes. This study shows that high classifica

关键词： Training data Image color analysis Feature extraction Vectors Skin cancer Lesions Task analysis Medical diagnosis Genetic programming classification algorithms Detection algorithms Image classification Neural networks

来源：评论

学校读者我要写书评

暂无评论

An Adversarial Example Defense Algorithm for Intelligent Driving

引用

IEEE NETWORK 2024年第6期38卷 98-105页

作者： Lu, Jiazhong Wang, Chenli Huang, Yuanyuan Ding, Kangyi Liu, Xiaolei Chengdu Univ Informat Technol Xin Gu Ind Coll Sch Cybersecur Chengdu 610225 Peoples R China Adv Cryptog & Syst Secur Key Lab Sichuan Prov Chengdu 610225 Peoples R China SUGON Ind Control & Secur Ctr Chengdu 610225 Peoples R China China Acad Engn Phys Inst Comp Applicat Mianyang 621900 Sichuan Peoples R China

In terms of intelligent driving, the adversarial example of an attack against traffic signs will cause the vehicle to make wrong judgments and decisions. However, the existing adversarial examples of defense algorithms generally have problems such as high training costs and poor defense effects and struggle to adapt to the environment of intelligent driving. In order to reduce the training cost while improving the accuracy of example classification, we propose a novel defense algorithm for adversarial examples combining micro-network structure and a generative adversarial network (GAN). The algorithm compresses the classification model and the Discriminator. And the Generator is designed to make the reconstructed sample generated closer to the real example distribution so as to solve the common problems of the existing adversarial example defense algorithm. Experiments on the collected traffic sign data set show that the proposed algorithm can achieve a better defense effect on the premise of lower training costs. The example classification accuracy can reach more than 97.9%, and the similarity between the reconstructed samples and the real examples reaches 96.26%. Moreover, the number of computations and parameters for training a single example is far lower than that of other commonly used defense methods, and the response speed is approximately doubled, which can greatly improve the safety of intelligent driving.

关键词： Training classification algorithms Generative adversarial networks Costs Computational modeling Generators Image reconstruction Intelligent driving Adversarial examples GAN Micro network structure

来源：评论

学校读者我要写书评

暂无评论

WMSNs Data Hidden Anomaly Detection Based on OH-KFJLT Bloom Filter

引用

IEEE ACCESS 2024年 12卷 174519-174526页

作者： Xiao, Chenkai Li, Zhongsheng Wu, Yanming Minxi Vocat & Tech Coll Informat Engn Dept Longyan 364021 Peoples R China Tsinghua Univ Sch Environm Beijing 100084 Peoples R China Zijin Intelligent Automat Xiamen Technol Co Ltd Xiamen 361026 Peoples R China

With the rapid development of the Industrial Internet of Things (IIoT), the volume and size of data handled by Wireless Multimedia Sensor Networks(WMSNs) have increased dramatically. Sensors suffer from damage due to continuous high-load usage and wear, leading to anomalies in sensor data collected and recorded. To solve the problem, this paper proposes an anomaly detection algorithm that is based on the Bloom Filter model, combined with the Optimal Hyperplane-based Kronecker Fast Johnson-Lindenstrauss Transform (OH-KFJLT) mapping and Reciprocal Competition Strategy(RCS), called Optimal Hyperplane Kronecker Fast Johnson-Lindenstrauss Transform Bloom Filter Anomaly Detection(OFBFAD). Firstly, the data is hashed by using the OH-KFJLT mapping method based on optimal hyperplane. Then, the RCS is adopted to de-noise the data. Finally, the Bloom Filter is constructed by 0-1 coding. In the simulation experiments conducted on the NAB, SMD, and COCO benchmark datasets, the False Alarm Rate (FAR) of the OFBFAD algorithm is all lower than 5%. The experimental results show that the OFBFAD has a higher Detection Rate (DR) and lower FAR than the current mainstream anomaly detection algorithm, and can be effectively applied to the anomaly detection of WMSNs data.

关键词： Anomaly detection Filtering algorithms Feature extraction classification algorithms Codes Encoding Industrial Internet of Things Sparse matrices Atmospheric modeling Wireless sensor networks Industrial Internet of Things (IIoT) wireless multimedia sensor networks (WMSNs) anomaly detection optimal hyperplane-based Kronecker Fast Johnson-Lindenstrauss transform (OH-KFJLT) reciprocal competition strategy(RCS)

来源：评论

学校读者我要写书评

暂无评论

A Unified Framework for Bandit Online Multiclass Prediction

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2025年第5期37卷 2200-2211页

作者： Feng, Wanjin Gao, Xingyu Zhao, Peilin Hoi, Steven C. H. Chinese Acad Sci Inst Microelect Beijing 100029 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Tencent AI Lab Shenzhen 518000 Peoples R China Singapore Management Univ Sch Comp & Informat Syst Singapore 188065 Singapore

Bandit online multiclass prediction plays an important role in many real-world applications. In this paper, we propose a unified Bandit Online Multiclass Prediction (BOMP) framework. This framework is based on our proposed margin-based gradient descent approach. Its update step provides an unbiased estimate of the surrogate loss gradient and has a lower variance than existing methods. It also enables our algorithms to update even for incorrect predictions by penalizing the wrong classes. The link function of the framework can evolve over time, gradually incorporating online data information including second-order information into the potential functions. Based on the proposed framework, we investigate first-order and second-order bandit online multiclass prediction algorithms. Theoretical analysis demonstrates the superiority of our proposed update rule and bandit online multiclass prediction framework. Finally, we compare our proposed first-order and second-order bandit online multiclass prediction algorithms with several state-of-the-art methods on two synthetic and four real-world datasets. The encouraging results show that our proposed algorithms significantly outperform state-of-the-art techniques.

关键词： Prediction algorithms Fasteners Heuristic algorithms classification algorithms Machine learning algorithms Logistics Vectors Bandit online learning margin-based update rule multiclass prediction unbiased loss estimator

来源：评论

学校读者我要写书评

暂无评论

The Power of Two Matrices in Spectral algorithms for Community Recovery

引用

IEEE TRANSACTIONS ON INFORMATION THEORY 2024年第5期70卷 3599-3621页

作者： Dhara, Souvik Gaudio, Julia Mossel, Elchanan Sandon, Colin Purdue Univ Sch Ind Engn W Lafayette IN 47907 USA Northwestern Univ Dept Ind Engn & Management Sci Evanston IL 60208 USA Massachussetts Inst Technol Dept Math Cambridge MA 02139 USA Ecole Polytech Fed Lausanne Chair Math Data Sci CH-1015 Lausanne Switzerland

Spectral algorithms are some of the main tools in optimization and inference problems on graphs. Typically, the graph is encoded as a matrix and eigenvectors and eigenvalues of the matrix are then used to solve the given graph problem. Spectral algorithms have been successfully used for graph partitioning, hidden clique recovery and graph coloring. In this paper, we study the power of spectral algorithms using two matrices in a graph partitioning problem. We use two different matrices resulting from two different encodings of the same graph and then combine the spectral information coming from these two matrices. We analyze a two-matrix spectral algorithm for the problem of identifying latent community structure in large random graphs. In particular, we consider the problem of recovering community assignments exactly in the censored stochastic block model, where each edge status is revealed independently with some probability. We show that spectral algorithms based on two matrices are optimal and succeed in recovering communities up to the information theoretic threshold. Further, we show that for most choices of the parameters, any spectral algorithm based on one matrix is suboptimal. The latter observation is in contrast to our prior works (2022a, 2022b) which showed that for the symmetric Stochastic Block Model and the Planted Dense Subgraph problem, a spectral algorithm based on one matrix achieves the information theoretic threshold. We additionally provide more general geometric conditions for the (sub)-optimality of spectral algorithms.

关键词： Encoding Stochastic processes Partitioning algorithms Symmetric matrices classification algorithms Clustering algorithms Optimization Stochastic block model spectral algorithms information-theoretic boundary

来源：评论

学校读者我要写书评

暂无评论

Feature Selection With Discernibility and Independence Criteria

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第11期36卷 6195-6209页

作者： Xie, Juanying Wang, Mingzhao Grant, Philip W. Pedrycz, Witold Shaanxi Normal Univ Sch Comp Sci Xian 710119 Peoples R China Swansea Univ Dept Comp Sci Swansea SA2 8PP Wales Univ Alberta Dept Elect & Comp Engn Edmonton AB T6G 2R3 Canada Macau Univ Sci & Technol Macau Inst Syst Engn Taipa Macau Peoples R China Istinye Univ Res Ctr Performance & Prod Anal TR-34010 Istanbul Turkiye

Feature selection plays a significant role in data mining and machine learning. It is challenging to determine how many features are necessary to form an optimal feature subset. To address this challenge, an innovative visual 2D feature selection framework is introduced, in which the feature discernibility and independence are defined to evaluate its capability for classification and its relevance to other features, respectively. All features are represented in 2D space with discernibility as $x$x-axis and independence as $y$y-axis. The features located in the upper right corner represent high discernibility and high independence, so comprise the optimal feature subset. This leads to the formation of a family of feature selection algorithms. Three such algorithms are proposed in this paper referred to as FSDIE, FSDIR, and FSDIS (Feature Selection based on the Discernibility and the Independence, respectively, of Exponent, Reciprocal, and anti-Similarity). To speed-up these three algorithms, a clustering based feature preselection first eliminates some unrelated and redundant features. Extensive experiments on UCI datasets, face datasets and gene expression datasets demonstrate that these three 2D feature selection algorithms are superior to the state-of-the-art methods indicating the power of our 2D feature selection framework.

关键词： Feature extraction classification algorithms Clustering algorithms Correlation Heuristic algorithms Redundancy Visualization Feature selection feature discernibility feature independence feature score 2D space

来源：评论

学校读者我要写书评

暂无评论

Measure-Driven Neural Solver for Optimal Transport Mapping

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第10期34卷 10437-10448页

作者： Li, Shenghao Li, Zezeng Wang, Zhanpeng Xu, Zebin Lei, Na Luo, Zhongxuan Dalian Univ Technol DUT Sch Software Technol Dalian 116024 Peoples R China Dalian Univ Technol DUT RU Int Sch Informat Sci & Engn Dalian 116024 Peoples R China

Optimal transport (OT) studies the most economical transformation of one probability measure into another, attracting attention across diverse fields and inspiring various OT-solving algorithms. However, adjusting the probability measure according to specific application requirements, such as achieving unbiased generated images or generating images with specific attributes, necessitates recalculating the OT mapping. This process may result in inefficiency and limited usage flexibility of existing algorithms. To address this, we propose a measure-driven neural solver for OT, the key of which is to construct a network module to learn Brenier's height representation, and then compute the gradient of Brenier's potential to derive the OT mapping. Our algorithm has two main advantages: i) It enables direct calculation or fine-tuning of the OT mapping when the target sample measure changes, enhancing efficiency. ii) For unbiased image generation or attribute-specific face generation, adjusting the posterior probability measure of the latent space in the pre-trained model suffices, without the need for additional auxiliary components, this highlights the flexibility of our algorithm. Extensive experiments demonstrate the excellent performance of our algorithm in debiased generation and controllable generation, and its flexibility and efficiency. In addition, both of these generation ways can enhance the classification performance of minority groups.

关键词： Task analysis classification algorithms Circuits and systems Costs Neural networks Data models Computational modeling Optimal transport Brenier's height representation debiased and controllable generation flexibility

来源：评论

学校读者我要写书评

暂无评论

Improved XGBoost and GM UWB/MEME IMU Positioning Methods for Non-Line-of-Sight Environments

引用

IEEE SENSORS JOURNAL 2024年第24期24卷 42384-42393页

作者： Sui, Xin Liao, Bangwen Wang, Changqiang Shi, Zhengxu Liaoning Tech Univ Sch Geomat Fuxin 123000 Peoples R China

The non-line-of-sight (NLOS) in ultrawide-band (UWB) wireless positioning systems in complex indoor environments is problematic. Therefore, in this study, a hybrid algorithm based on particle swarm optimization (PSO)-genetic algorithm (GA) is proposed to optimize the NLOS recognition algorithm with extreme gradient boosting (XGBoost), and the gray prediction model (GM) is then used to correct the NLOS data. A classification feature system is constructed based on UWB data for XGBoost decision trees. On this basis, the gray relational analysis (GRA) algorithm is used to calculate the comprehensive importance measurement index of each feature, and these features are subsequently used to construct an effective feature subset and improve the accuracy of the NLOS recognition. The PSO algorithm has memory and tends to produce local optimal solutions, whereas the GA algorithm has good global convergence but no memory for individuals. Therefore, to prevent the PSO algorithm from falling into local optima, a hybrid algorithm, the PSO-GA, is adopted to optimize the parameters of the XGBoost model. The optimized XGBoost model is used for NLOS recognition in UWB systems and further improves the accuracy of NLOS recognition. For UWB data identified as NLOS data, the GM is used for correction to improve the utilization of the UWB measurement values. Then, the corrected UWB data are combined with the microelectromechanical system (MEMS) inertial measurement unit (IMU) using a tight coupling method to achieve combined positioning, which avoids possible multipath interference or signal attenuation problems when UWB positioning is used alone. The experimental results show that in complex indoor environments, the proposed algorithm can effectively recognize and correct the NLOS signals. The root mean square errors of the combined UWB/MEMS IMU positioning system in the X- and Y-directions are 0.107 and 0.076 m, respectively. Compared with the original UWB data and MEMS IMU combination r

关键词： Accuracy Feature extraction classification algorithms Distance measurement Correlation Root mean square Prediction algorithms Optimization Micromechanical devices Indexes Extreme gradient boosting (XGBoost) indoor positioning non-line-of-sight (NLOS) error particle swarm optimization (PSO) genetic hybrid algorithm ultrawide-band (UWB) microelectromechanical system inertial measurement unit (MEMS IMU) combination

来源：评论

学校读者我要写书评

暂无评论

Identifying Expert Behavior in Offline Training Datasets Improves Behavioral Cloning of Robotic Manipulation Policies

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第2期9卷 1294-1301页

作者： Wang, Qiang McCarthy, Robert Bulens, David Cordova Sanchez, Francisco Roldan McGuinness, Kevin O'Connor, Noel E. Redmond, Stephen J. Univ Coll Dublin Sch Elect & Elect Engn Dublin D14 PF57 Ireland Univ Coll Dublin CeADAR Ireland Ctr Appl AI Dublin D14 PF57 Ireland Dublin City Univ Sch Elect Engn Dublin D9 Ireland

This letter presents our solution for the Real Robot Challenge III1, aiming to address dexterous robotic manipulation tasks through learning from offline data. In this competition, participants were given two types of datasets for each task: expert and mixed. Each expert dataset is collected by a high-skill policy, whereas the mixed dataset is collected using both expert and non-expert policies. We found that the vanilla behavioural cloning (BC) can learn a very proficient policy with minimal human intervention when trained on expert datasets. Notably, BC outperformed even the most advanced offline reinforcement learning (RL) algorithms. However, when applied to mixed datasets, the performance of BC deteriorates;similarly, the performance of offline RL algorithms is also less than satisfactory. Upon examining the provided datasets, it was apparent that each mixed dataset contained a significant proportion of expert data, which should enable the training of a proficient BC agent. However, the expert data is not labelled in the datasets. As a result, we propose a classifier to identify the pattern of the expert behaviour within a mixed dataset and then utilize it to isolate the expert data. To further boost the BC performance, we take advantage of the geometric symmetry of the arena to augment the training dataset through mathematical transformations. Ultimately, our submission outperformed that of all other participants.

关键词： Robots Task analysis Robot kinematics Training Behavioral sciences classification algorithms Cloning Data sets for robot learning imitation learning reinforcement 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：