检索结果-内蒙古大学图书馆

Personalized Federated Lithium-Ion Battery Capacity prediction via Cluster and Fusion Modules

IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2024年第3期10卷 6434-6448页

作者： Xiao, Fei Wu, Lifeng Capital Normal Univ Informat Engn Coll Beijing 100048 Peoples R China Capital Normal Univ Beijing Key Lab Elect Syst Reliabil Technol Beijing 100048 Peoples R China

Federated learning (FL) is a promising solution for addressing information security sharing challenges in the Internet of Vehicles (IoV). It enables individual-level capacity prediction of lithium-ion batteries in electric vehicles (EVs). However, existing FL algorithms primarily focus on training a single global sharing model, neglecting the predictive capabilities of individual participants, which poses challenges for batteries with distinct capacity degradation trends. To address this limitation, we introduce battery-personalized FL (BT-PFL), a novel FL framework that provides personalized capacity prediction models based on the local data distribution of each battery. Our approach involves constructing cluster and fusion modules, creating a personalized learning space, and leveraging valuable domain knowledge. On the one hand, we proactively identify the capacity degradation distribution of different batteries, ensuring data privacy during FL within clusters. Each cluster retains domain-specific style information, enabling collaborative training and shared model parameters among participants. On the other hand, we introduce a knowledge distillation (KD) algorithm that facilitates knowledge transfer between clusters by constructing a teacher model incorporating multidomain knowledge. The experimental results on both single-domain and multidomain show that our approach not only significantly enhances individual battery prediction accuracy but also outperforms other methods across various datasets, network structures, and noise levels.

关键词： Data models Lithium-ion batteries Predictive models Training prediction algorithms Transportation Federated learning Capacity prediction data heterogeneity federated learning (FL) lithium-ion battery long short-term memory (LSTM)

来源：评论

学校读者我要写书评

暂无评论

Posts Quality prediction for StackOverflow Website

引用

IEEE ACCESS 2024年 12卷 135601-135615页

作者： Hu, Jiawei Yang, Bo Beijing Forestry Univ Sch Informat Sci & Technol Beijing 100083 Peoples R China

The development of the computer industry is closely linked to various question-and-answer websites, whose primary function is to discover and solve problems encountered by users. This paper focuses on the quality prediction of question posts on the StackOverflow(SO) website, which can essentially be considered a text classification problem. Given the large number of users, manual moderation becomes inadequate when faced with a vast quantity of user questions. Reducing the occurrence of low-quality questions can effectively alleviate the operational pressure on the website. We preprocess and vectorize the posts to obtain vector representations of the training and testing sets. After training 5 different machine learning models, including decision trees, random forests, naive Bayes, support vector machines, logistic regression, and 2 deep learning models, Bi-LSTM and BERT, these models are compared through experiments by adjusting the values of different parameters. The results indicate that different parameters have a certain impact on the experimental results, and there are significant differences in the quality prediction performance of different models. The lowest accuracy rate only reaches 54%, while the highest accuracy is 92%. The comparison shows that quality assessment based on the attention mechanism model is effective and can be used to predict post-quality.

关键词： Accuracy Training Feature extraction Random forests prediction algorithms Machine learning Quality assessment Deep learning Text categorization Quality-analysis deep-learning machine-learning NLP text classification

来源：评论

学校读者我要写书评

暂无评论

Perception-Aware Texture Similarity prediction

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2024年 33卷 3536-3549页

作者： Wang, Weibo Dong, Xinghui Ocean Univ China Sch Comp Sci & Technol Qingdao 266100 Peoples R China

Texture similarity plays important roles in texture analysis and material recognition. However, perceptually-consistent fine-grained texture similarity prediction is still challenging. The discrepancy between the texture similarity data obtained using algorithms and human visual perception has been demonstrated. This dilemma is normally attributed to the texture representation and similarity metric utilised by the algorithms, which are inconsistent with human perception. To address this challenge, we introduce a Perception-Aware Texture Similarity prediction Network (PATSP-Net). This network comprises a Bilinear Lateral Attention Transformer network (BiLAViT) and a novel loss function, namely, RSLoss. The BiLAViT contains a Siamese Feature Extraction Subnetwork (SFEN) and a Metric Learning Subnetwork (MLN), designed on top of the mechanisms of human perception. On the other hand, the RSLoss measures both the ranking and the scaling differences. To our knowledge, either the BiLAViT or the RSLoss has not been explored for texture similarity tasks. The PATSP-Net performs better than, or at least comparably to, its counterparts on three data sets for different fine-grained texture similarity prediction tasks. We believe that this promising result should be due to the joint utilization of the BiLAViT and RSLoss, which is able to learn the perception-aware texture representation and similarity metric.

关键词： Feature extraction Measurement Task analysis Transformers prediction algorithms Convolutional neural networks Computational modeling Texture similarity texture retrieval texture synthesis quality assessment bilinear attention ranking loss

来源：评论

学校读者我要写书评

暂无评论

A Cooperative Vehicle Localization and Trajectory prediction Framework Based on Belief Propagation and Transformer Model

引用

IEEE TRANSACTIONS ON CONSUMER ELECTRONICS 2024年第1期70卷 2746-2758页

作者： Jin, Feiyu Liu, Kai Liu, Chunhui Cheng, Tongtong Zhang, Hao Lee, Victor C. S. Chongqing Univ Coll Comp Sci Chongqing 400040 Peoples R China Chongqing Univ Posts & Telecommun Coll Comp Sci & Technol Chongqing 400065 Peoples R China Univ Hong Kong Dept Elect & Elect Engn Hong Kong Peoples R China

The advancement of sensing, transmission, and computation technologies has transformed modern vehicles into ubiquitous consumer electronics. This paper presents a cooperative vehicle localization and trajectory prediction framework for enabling Intelligent Transportation Systems (ITS) applications. Specifically, the proposed framework consists of a Belief Propagation based Location Approximation (BPLA) algorithm for cooperative vehicle localization and a Transformer-based Vehicle trajectory prediction model called VFormer. The BPLA algorithm first establishes a factor graph based on the sensor measurements transmitted by vehicles, and then adopts a modified belief propagation procedure to approximate the posterior distribution of vehicles. On this basis, VFormer extracts the hidden features from historical positions estimated by BPLA and vehicle motion data to model long-term and short-term motion patterns of vehicles. Moreover, the multi-head attention layer in the VFormer is utilized to learn the spatial-temporal dependencies between the target vehicle and its surrounding vehicles at different timestamps to improve prediction accuracy. A comprehensive performance evaluation has been conducted based on the public vehicle trajectory dataset and the real-world system prototype. Experiment results demonstrate the superiority of the proposed framework on improving vehicle localization and trajectory prediction performance.

关键词： Location awareness Trajectory Predictive models Global navigation satellite system Belief propagation prediction algorithms Consumer electronics Vehicle localization trajectory prediction belief propagation transformer model Internet of Vehicles

来源：评论

学校读者我要写书评

暂无评论

Construction of Dynamic Traffic Pattern Recognition and prediction Model Based on Deep Learning in the Background of Intelligent Cities

引用

IEEE ACCESS 2024年 12卷 1418-1433页

作者： Cui, Jian Zhao, Jianyou Changan Univ Sch Automobile Xian 710064 Peoples R China

To reduce traffic congestion, it is particularly important to use advanced technology to predict urban traffic flow. Therefore, a dynamic traffic pattern prediction model is proposed, which includes convolutional neural network, long and short term memory network and attention mechanism. The validity of the prediction model is verified by the loss function and the average absolute percentage error. In addition, the study also constructs a model for user travel pattern and parking point recognition based on deep learning and mobile signaling data. The performance of the recognition model is verified by the accuracy and other indicators. The research outcomes demonstrated that the max average absolute percentage error of the dynamic traffic mode prediction model was 7.8%, and the mini value was 2.9%. The average accuracy of the user travel pattern recognition model was 83.34%, and that of the parking point recognition model was 88.56%.The dynamic traffic model recognition and prediction model designed by the research institute has better results, and has practical guiding significance in smart city traffic management.

关键词： Predictive models Pattern recognition Clustering algorithms Data models Traffic control prediction algorithms Roads Traffic mode recognition prediction mobile signaling data attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Future Air Quality prediction Using Long Short-Term Memory Based on Hyper Heuristic Multi-Chain Model

引用

IEEE ACCESS 2024年 12卷 123678-123693页

作者： Chatterjee, Kalyan Kumar, Samla Suraj Kumar, Ramagiri Praveen Bandyopadhyay, Anjan Swain, Sujata Mallik, Saurav Al-Rasheed, Amal Abbas, Mohamed Soufiene, Ben Othman Nalla Malla Reddy Engn Coll Dept Comp Sci & Engn Hyderabad 500088 Telangana India Nalla Malla Reddy Engn Coll Dept Artificial Intelligence & Machine Learning Hyderabad 500088 India Kalinga Inst Ind Technol Sch Comp Sci Engn Bhubaneswar 751024 Odisha India Harvard TH Chan Sch Publ Hlth Dept Environm Hlth Boston MA 02115 USA Princess Nourah Bint Abdulrahman Univ Coll Comp & Informat Sci Dept Informat Syst POB 84428 Riyadh 11671 Saudi Arabia King Khalid Univ Coll Engn Elect Engn Dept Abha 61421 Saudi Arabia Univ Sousse PRINCE Lab Res ISITcom Sousse 4023 Tunisia

Air pollution is a critical global concern, demanding precise air quality forecasting to mitigateits severe consequences. Our study introduces Future Air Quality prediction using Long Short-TermMemory based on Hyper Heuristic Multi-Chain Model (H2MCM) to project future air quality, consideringvarious meteorological factors (MFs) and pollution-related variables like atmospheric pressure, temperature,humidity, and wind patterns. Leveraging 12 units of Long Short-Term Memory neural networks (LSTMs),H2MCM accurately predicts forthcoming air pollutants (APs) concentrations such as particulate matter withdiameter 2.5 mu m (PM2.5), carbon monoxide (CO), and nitrogen dioxide (NO2). Additionally, it accountsfor spatiotemporal correlations between these APs and MFs, which significantly influence the air qualityprediction for the next immediate time interval. H2MCM utilizes a multi-chain mechanism, employing1-hour prediction models to forecast air quality hourly, enabling approximations for the next 12 ***, for an efficient model selection, Akaike Information Criterion (AIC), Schwarz Bayesian InformationCriterion (SBIC), Hannan-Quinn Information Criterion (HQIC), and corrected AIC (AICc) tools are usedbased on their ability to balance model fit and complexity. Furthermore, it demonstrates the ability toenhance the performance of any predictor. Experimental results substantiate H2MCM's superiority overvarious models, including the Support Vector Regressor (SVR), Multi-Layer Perceptron (MLP), RecurrentAir Quality Predictor (RAQP), and Valchogianni models. H2MCM achieves impressive up to 75% betteraccuracy and consistency compared to SVR, 60% better than MLP, 38% better than RAQP, and 70% betterthan Valchogianni models.

关键词： Atmospheric modeling Predictive models Accuracy Computational modeling Air pollution Data models prediction algorithms Air quality Heuristic algorithms air pollutant concentrations (APCs) deep learning (DL) heuristic machine learning (ML) meteorological factors (MFs) multi-chain regressors

来源：评论

学校读者我要写书评

暂无评论

Mitigating Decision Boundary Confusion: A Classifier prediction-Oriented Domain Adaptation Network

引用

IEEE SENSORS JOURNAL 2024年第11期24卷 18672-18684页

作者： Ding, Xue Deng, Aidong Deng, Minqiang Liu, Dongying Shi, Yaowei Southeast Univ Sch Energy & Environm Natl Engn Res Ctr Power Generat Control & Safety Nanjing 210096 Peoples R China Harbin Turbine Co Ltd Harbin 150046 Peoples R China Nanjing Univ Posts & Telecommun Sch Internet Things Nanjing 210003 Peoples R China

Domain adaptation has been extensively studied and applied in fault diagnosis of rotating machinery to address significant distribution differences between source and target domains. In industrial production, however, decision boundary confusion caused by strong connectivity among similar categories is prevalent. In response to this challenge, a classifier prediction-oriented domain adaptation network (CPDAN) is proposed in this article. Specifically, CPDAN contains two parts. Firstly, a target domain category separation algorithm (TDCSA) is designed to address the problem of decision boundary confusion. TDCSA effectively reduces intercategory correlations by minimizing the inner product between class predictions and their corresponding transposes within the target domain. It enhances feature discriminability, thereby facilitating category separation. Meanwhile, a similar uncertainty mechanism (SUM) is developed to highlight samples with high contributions to category separation. Secondly, to enhance the application effectiveness of TDCSA, a class feature distribution alignment algorithm (CFDAA) oriented to classifier prediction feature reinforcement mechanism (CPFRM) is proposed. CFDAA augments the generated features based on classifier-predicted class information and embeds these new features into a distance regularization term to guide class-level distribution alignment. This effectively captures feature transferability, further strengthening interdomain similarity. Extensive experiments on two bearing datasets have demonstrated the effectiveness and superiority of CPDAN.

关键词： Feature extraction Sensors Fault diagnosis Neurons Machinery Classification algorithms prediction algorithms Category separation classifier prediction decision boundary confusion domain adaptation feature discriminability feature transferability

来源：评论

学校读者我要写书评

暂无评论

Judgment prediction Based on Tensor Decomposition With Optimized Neural Networks

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024年第8期35卷 11116-11127页

作者： Guo, Xiaoding Zhang, Lei Tian, Zhihong Henan Univ Sch Comp & Informat Engn Kaifeng 450001 Peoples R China Henan Univ Henan Key Lab Big Data Anal & Proc Kaifeng 450001 Peoples R China Guangzhou Univ Cyberspace Inst Adv Technol Guangzhou 510006 Peoples R China

In the field of smart justice, handling legal cases through artificial intelligence technology is a research hotspot. Traditional judgment prediction methods are mainly based on feature models and classification algorithms. The former is difficult to describe cases from multiple angles and capture the correlation information between different case modules, while requires a wealth of legal expertise and manual labeling. The latter is unable to accurately extract the most useful information from case documents and produce fine-grained predictions. This article proposes a judgment prediction method based on tensor decomposition with optimized neural networks, which consists of OTenr, GTend, and RnEla. OTenr represents cases as normalized tensors. GTend decomposes normalized tensors into core tensors using the guidance tensor. RnEla intervenes in a case modeling process in GTend by optimizing the guidance tensor, so that core tensors represent tensor structural and elemental information, which is most conducive to improving the accuracy of judgment prediction. RnEla consists of the similarity correlation Bi-LSTM and optimized Elastic-Net regression. RnEla takes the similarity between cases as an important factor for judgment prediction. Experimental results on real legal case dataset show that the accuracy of our method is higher than that of the previous judgment prediction methods.

关键词： Tensors Predictive models prediction algorithms Law Classification algorithms Vocabulary Correlation Judgment prediction legal cases Index Terms neural networks tensor decomposition

来源：评论

学校读者我要写书评

暂无评论

Efficiency Improvement to Neural-Network-Driven Optimal Path Planning via Region and Guideline prediction

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第2期9卷 1851-1858页

作者： Huang, Yuan Tsao, Cheng-Tien Lee, Hee-Hyol Univ Waseda Grad Sch Informat Prod & Syst Kitakyushu 8080135 Japan

Traditional sampling-based algorithms rely on random samples to explore a whole configuration space of robots for optimal path planning, while a uniform sampler impedes the exploration with randomly generated samples, leading to a long calculation time, especially in complex environments. Recently, neural-network-driven methods have attracted wide interest in developing non-uniform sampling to improve the sampling efficiency and reduce the calculation time. A region that contains an optimal path is predicted by neural networks and employed subsequently to biasedly generate samples. This work aims at enhancing the sampling efficiency and reducing the calculation time of the optimal path planning by a novel region and guideline prediction (denoted as RGP) model. We innovatively propose the RGP model with a guideline prediction module to estimate the guideline distributions, which are characterized by the central line of the predicted region. The predicted region and guideline are integrated into a sampling-based algorithm, namely RGP-RRT*, with an adaptively biased sampling strategy to select a proper domain for sampling. Simulations demonstrate the RGP model outperforms other region prediction models in accuracy and robustness. Besides, the RGP-RRT* reliably achieves a 7.2-80.1% reduction in calculation time and a 2.0-58.1% reduction in sample number compared with other neural-network-driven methods.

关键词： Path planning Guidelines Predictive models prediction algorithms Convolution Adaptation models Space exploration Motion and path planning AI-based methods

来源：评论

学校读者我要写书评

暂无评论

GFS-Node: Graph Fuzzy Systems for Node prediction

引用

IEEE TRANSACTIONS ON FUZZY SYSTEMS 2024年第12期32卷 6822-6834页

作者： Hu, Fuping Deng, Zhaohong Xie, Zhenping Zhang, Te Choi, Kup-Sze Zhang, Fan Wang, Shitong Jiangnan Univ Sch Artificial Intelligence & Comp Sci Wuxi 214122 Peoples R China Minist Educ Engn Res Ctr Intelligent Technol Healthcare Wuxi 214122 Peoples R China Univ Nottingham Sch Comp Sci Lab Uncertainty Data & Decis Making Nottingham NG7 2RD England TechCosmos Ltd Hong Kong Peoples R China Inspur Zhuoshu Big Data Ind Dev Co Ltd Wuxi 214062 Peoples R China

Graph data modeling is nontrivial due to the challenges to ensure model interpretability and handle data uncertainty. While methods derived from deep learning models, such as graph neural networks (GNNs), are able to handle graph data, the interpretability is limited. Graph fuzzy systems (GFSs) based on the fuzzy rules and fuzzy inference have been proposed to improve interpretability, but the existing methods are developed for whole graph prediction only and cannot deal with node prediction, which is a more common task in graph data modeling. To tackle the challenges, a novel GFS for node prediction (GFS-node) is investigated in this study. For this purpose, the concepts, framework, and algorithms of GFS-node are systematically developed. First, several related concepts are defined, including the node fuzzy rule base, node fuzzy set, and node consequent processing module (NCPM). A general framework for GFS-node is then presented, where the construction of antecedents and the consequents of fuzzy rules are analyzed. Furthermore, a concrete implementation method of GFS-node is designed. In particular, the kernel K virtual central nodes clustering (KVCN) algorithm is proposed to develop the algorithm for antecedent generation, and the linear message passing network (LMPN) is adopted to develop the algorithm for consequent generation and learning. Experiments are carried out on multiple benchmark datasets, and the results show that GFS-node combines the advantages of both traditional fuzzy systems and classical GNNs for node prediction.

关键词： Fuzzy systems Data models Predictive models Uncertainty prediction algorithms Fuzzy sets Clustering algorithms Graph fuzzy systems for node modeling (GFS-node) graph learning kernel K virtual central nodes clustering (KVCN) linear message passing network (LMPN) node consequent processing module (NCPM) node fuzzy sets node prediction

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：