检索结果-南通市图书馆

A credible traffic prediction method based on self-supervised causal discovery

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第5期67卷 244-258页

作者： Dan WANG Yingjie LIU Bin SONG State Key Laboratory of Integrated services Networks Xidian University

Next-generation wireless network aims to support low-latency, high-speed data transmission services by incorporating artificial intelligence(ai) technologies. To fulfill this promise, ai-based network traffic prediction is essential for pre-allocating resources, such as bandwidth and computing power. This can help reduce network congestion and improve the quality of service(QoS) for users. Most studies achieve future traffic prediction by exploiting deep learning and reinforcement learning, to mine spatio-temporal correlated variables. Nevertheless, the prediction results obtained only by the spatio-temporal correlated variables cannot reflect real traffic changes. This phenomenon prevents the true prediction variables from being inferred, making the prediction algorithm perform poorly. Inspired by causal science, we propose a novel network traffic prediction method based on self-supervised spatio-temporal causal discovery(SSTCD). We first introduce the Granger causal discovery algorithm to build a causal graph among prediction variables and obtain spatio-temporal causality in the observed data, which reflects the real reasons affecting traffic ***, a graph neural network(GNN) is adopted to incorporate causality for traffic prediction. Furthermore,we propose a self-supervised method to implement causal discovery to to address the challenge of lacking ground-truth causal graphs in the observed data. Experimental results demonstrate the effectiveness of the SSTCD method.

关键词： wireless network traffic prediction causal discovery self-supervised

A Comprehensive Systematic Review: Advancements in Skin Cancer Classification and Segmentation Using the ISIC Dataset

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2024年第9期140卷 2131-2164页

作者： Madiha Hameed Aneela Zameer Muhammad Asif Zahoor Raja Department of Computer Sciences Pakistan Institute of Engineering and Applied SciencesNilore50700Islamabad Future Technology Research Center National Yunlin University of Science and TechnologyDouliou64002TaiwanChina Directorate of IT services University of GujratGujrat50700Pakistan

The International Skin Imaging Collaboration(ISIC)datasets are pivotal resources for researchers in machine learning for medical image analysis,especially in skin cancer *** datasets contain tens of thousands of dermoscopic photographs,each accompanied by gold-standard lesion diagnosis *** challenges associated with ISIC datasets have spurred significant advancements,with research papers reporting metrics surpassing those of human *** cancers are categorized into melanoma and non-melanoma types,with melanoma posing a greater threat due to its rapid potential for metastasis if left *** paper aims to address challenges in skin cancer detection via visual inspection and manual examination of skin lesion images,processes historically known for their *** notable advancements in machine learning and deep learning models,persistent challenges remain,largely due to the intricate nature of skin lesion *** review research on convolutional neural networks(CNNs)in skin cancer classification and segmentation,identifying issues like data duplication and augmentation *** explore the efficacy of Vision Transformers(ViTs)in overcoming these challenges within ISIC dataset *** leverage their capabilities to capture both global and local relationships within images,reducing data duplication and enhancing model ***,ViTs alleviate augmentation issues by effectively leveraging original *** a thorough examination of ViT-based methodologies,we illustrate their pivotal role in enhancing ISIC image classification and *** study offers valuable insights for researchers and practitioners looking to utilize ViTs for improved analysis of dermatological ***,this paper emphasizes the crucial role of mathematical and computational modeling processes in advancing skin cancer detection methodologies,highlighting their significance in improving algorithmic

关键词： Medical image skin cancer classification skin cancer segmentation international skin imaging collaboration convolutional neural network deep learning

A novel approach towards parametric assessment of reliability and resilience of high voltage mica-based insulation systems by statistical analysis of experimental failure data

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

High Voltage 2024年第2期9卷 495-507页

作者： Shahram Negari Davoud Esmaeil Moghadam Integrated Power services ClevelandOhioUSA Valeo eAutomotive Germany GmbH ErlangenGermany

Insulation systems in high-voltage electric machines play a pivotal role in the reliable operation and longevity of the ***-based insulation materials have proven to possess and maintain excellent dielectric properties in the long run and prevent premature insulation *** qualifications tests,such as voltage endurance,are outlined in IEC and IEEE *** authors,however,take a different parametric approach,opting for reliability assessment of insulation systems using derived three-parameter Weibull ***,instead of simple pass–fail criteria,empirical data is employed to determine failure rate probabilities quantitatively and *** data,including breakdown,dissipation factor,and partial discharge mea-surements,are used to construct the Weibull distribution model to predict fault and failure rates and calculate hazard *** rigorous examinations interpreted through the analytical model help assess insulation system resilience and particularly the impact of electrical field stress and mica *** of electrical stress from 66.75 to 71.20 V/mil demonstrated how the mean time to failure of the system changed from 146.4 to 85.1 at 3 Un,hence identifying opportunities for design improvement and uncovering performance ***,the developed framework enhances comprehension of insulation system failure probabilities,guiding design decisions and ensuring a secure and reliable operation of electrical machines across applications.

关键词： insulation parametric approach

Mitigate noisy data for smart IoT via GAN based machine unlearning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第3期67卷 62-78页

作者： Zhuo MA Yilong YANG Yang LIU Xinjing LIU Jianfeng MA School of Cyber Engineering Xidian University State Key Laboratory of Integrated services Networks (ISN)

With the development of IoT applications, machine learning dramatically improves the utility of variable IoT systems such as autonomous driving. Although the pretrain-finetune framework can cope well with data heterogeneity in complex IoT scenarios, the data collected by sensors often contain unexpected noisy data, e.g., out-of-distribution(OOD) data, which leads to the reduced performance of fine-tuned models. To resolve the problem, this paper proposes Mu GAN, a method that can mitigate the side-effect of OOD data via the generative adversarial network(GAN)-based machine unlearning. Mu GAN follows a straightforward but effective idea to mitigate the performance loss caused by OOD data, i.e., “flashbacking” the model to the condition where OOD data are excluded from model training. To achieve the goal, we design an adversarial game, where a discriminator is trained to identify whether a sample belongs to the training set by observing the confidence score. Meanwhile, a generator(i.e., the target model) is updated to fool the discriminator into believing that the OOD data are not included in the training set but others do. The experimental results show that benefiting from the high unlearning rate(more than 90%) and retention rate(99%), Mu GAN succeeds in lowering the model performance degradation caused by OOD data from 5.88% to 0.8%.

关键词： machine unlearning generative adversarial network out of distribution data Internet of Thing neural network

Optimization study of spherical tuyere based on BP neural network and new evaluation index

同方期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Building Simulation 2024年第2期17卷 223-234页

作者： Mengchao Liu Ran Gao Yi Wang Angui Li School of Building services Science and Engineering Xi’an University of Architecture and TechnologyXi’an710055China

The energy consumption of heating,ventilation,and air conditioning(HVAC)systems holds a significant position in building energy usage,accounting for about 65%of the total energy ***,with the advancement of building automation,the energy consumption of ventilation systems continues to *** study focuses on improving the performance of spherical tuyeres in HVAC *** primarily utilizes neural networks and multi-island genetic algorithms(MIGA)for multi-parameter *** employing methods such as structural parameterization,accurate and fast computational fluid dynamics(CFD)simulations,a minimized sample space,and a rational optimization strategy,the time cycle of the optimization process is ***,a new comprehensive evaluation index is proposed in this research to describe the performance of spherical tuyeres,which can be used to more accurately assess spherical tuyeres with different *** results show that by establishing a neural network prediction model and combining it with the multi-island genetic algorithm,a novel spherical tuyere design was successfully *** optimized novel spherical tuyeres achieved a 27.05%reduction in the spherical tuyeres effective index(STEI)compared to the traditional spherical ***,the resistance decreased by 15.68%,and the jet length increased by 7.57%.The experimental results demonstrate that our proposed optimization method exhibits high accuracy,good generalization capability,and excellent agreement at different Reynolds numbers.

关键词： spherical tuyeres STEI CFD ventilation

Analysis of Maximum Liquid Carrying Capacity Based on Conventional Tubing Plunger Gas Lift

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Energy Engineering 2024年第6期121卷 1521-1533页

作者： Yanqun Yu Wenhao Xu Yahui Huangfu Jinhai Liu Bensheng Wang Kai Liu College of Mechanical and Electronic Engineering China University of Petroleum(East China)Qingdao266580China Energy Technology&services Oil Production services Company CNOOCTianjin300450China

China’s unconventional gas fields have a large number of low-productivity and low-efficiency wells, many of whichare located in remote and environmentally harsh mountainous areas. To address the long-term stable productionof these gas wells, plunger-lift technology plays an important role. In order to fully understand and accurately graspthe drainage and gas production mechanisms of plunger-lift, a mechanical model of plunger-liquid column uplift inthe plunger-lift process was established, focusing on conventional plunger-lift systems and representative wellboreconfigurations in the Linxing region. The operating casing pressure of the plunger-lift process and the calculationmethod for the maximum daily fluid production rate based on the work regime with the highest fluid recovery ratewere determined. For the first time, the critical flow rate method was proposed as a constraint for the maximumliquid-carrying capacity of the plunger-lift, and liquid-carrying capacity charts for conventional plunger-lift withdifferent casing sizes were developed. The results showed that for 23/8 casing plunger-lift, with a well depth ofshallower than 808 m, the maximum drainage rate was 33 m3/d;for 27/8 casing plunger-lift, with a well depth ofshallower than 742 m, the maximum drainage rate was 50.15 m3/d;for 31/2 casing plunger-lift, with a well depthof shallower than 560 m, the maximum drainage rate was 75.14 m3/d. This research provides a foundation for thescientific selection of plunger-lift technology and serves as a decision-making reference for developing reasonableplunger-lift work regimes.

关键词： Unconventional well conventional tubing string plunger gas lift critical flow rate liquid carry-over chart version

Experimental study on the CO_(2) concentration and age of air distribution inside tiny sleeping spaces

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Building Simulation 2024年第4期17卷 591-605页

作者： Jingying Zhang Yanyan Li Haiguo Yin Linfeng Liang Songmei Zu Le Gao Ying Zhang Angui Li School of Building services Science and Engineering Xi’an University of Architecture and TechnologyXi’anShaanxi710055China

In recent years,rapid urban development has led to capsule hotels,sleep pods,and other tiny sleeping spaces that adapt to people’s fast-paced lives,achieving maximum functionality with a very small ***,due to the small space,human metabolic pollutant(such as CO_(2))is more likely to accumulate,and the air is not easily *** this paper,a full-size experimental platform is set up with three types of ventilation modes to explore the exclusion efficiency of metabolic pollutants and the overall distribution of age of air under these ventilation *** conclusions showed that the mean values of metabolic pollutant exclusion rates for the different ventilation modalities varied very little across the spatial dimensions of the confined space but varied considerably in the area around the *** double-side attached ventilation method was the most effective in removing human metabolic pollutants,especially in the head region(CN≥0.92),while the single-wall attached ventilation method had the best air exchange efficiency(η≥0.85).This suggests an inconsistent distribution of CO_(2) and age of air,which is contrary to general common *** conclusions of this paper can guide the design of ventilation for tiny sleeping spaces.

关键词： attached ventilation mode human metabolic pollutant indoor air quality air exchange efficiency tiny sleeping spaces

GraphSTGAN:Situation understanding network of slow-fast high maneuvering targets for maritime monitor services of IoT data

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Digital Communications and Networks 2024年第3期10卷 620-630页

作者： Guanlin Wu Haipeng Wang Yu Liu You He Shenzhen International Graduate School Tsinghua UniversityShenzhen518071China Institute of Space Science and applied Technology Harbin Institute of TechnologyShenzhen518055China Department of Electronic Engineering Tsinghua UniversityBeijing100084China

With the rapid growth of the maritime Internet of Things(IoT)devices for Maritime Monitor services(MMS),maritime traffic controllers could not handle a massive amount of data in *** unmanned MMS,one of the key technologies is situation ***,the presence of slow-fast high maneuvering targets and track breakages due to radar blind zones make modeling the dynamics of marine multi-agents difficult,and pose significant challenges to maritime situation *** order to comprehend the situation accurately and thus offer unmanned MMS,it is crucial to model the complex dynamics of multi-agents using IoT big ***,previous methods typically rely on complex assumptions,are plagued by unstructured data,and disregard the interactions between multiple agents and the spatial-temporal correlations.A deep learning model,Graph Spatial-Temporal Generative Adversarial Network(GraphSTGAN),is proposed in this paper,which uses graph neural network to model unstructured data and uses STGAN to learn the spatial-temporal dependencies and *** experiments show the effectiveness and robustness of the proposed method.

关键词： Internet of things Multi-agents Graph neural network Maritime monitoring services

Deep Learning and Tensor-Based Multiple Clustering Approaches for Cyber-Physical-Social Applications

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第3期78卷 4109-4128页

作者： Hongjun Zhang Hao Zhang Yu Lei Hao Ye Peng Li Desheng Shi School of Computer Science Nanjing University of Posts and TelecommunicationsNanjing210023China China Communications services Co. Ltd.Beijing100071China China Communications services Hexin Science&Technology Co. Ltd.Hefei230031China

The study delves into the expanding role of network platforms in our daily lives, encompassing various mediums like blogs, forums, online chats, and prominent social media platforms such as Facebook, Twitter, and Instagram. While these platforms offer avenues for self-expression and community support, they concurrently harbor negative impacts, fostering antisocial behaviors like phishing, impersonation, hate speech, cyberbullying, cyberstalking, cyberterrorism, fake news propagation, spamming, and fraud. Notably, individuals also leverage these platforms to connect with authorities and seek aid during disasters. The overarching objective of this research is to address the dual nature of network platforms by proposing innovative methodologies aimed at enhancing their positive aspects and mitigating their negative repercussions. To achieve this, the study introduces a weight learning method grounded in multi-linear attribute ranking. This approach serves to evaluate the significance of attribute combinations across all feature spaces. Additionally, a novel clustering method based on tensors is proposed to elevate the quality of clustering while effectively distinguishing selected features. The methodology incorporates a weighted average similarity matrix and optionally integrates weighted Euclidean distance, contributing to a more nuanced understanding of attribute importance. The analysis of the proposed methods yields significant findings. The weight learning method proves instrumental in discerning the importance of attribute combinations, shedding light on key aspects within feature spaces. Simultaneously, the clustering method based on tensors exhibits improved efficacy in enhancing clustering quality and feature distinction. This not only advances our understanding of attribute importance but also paves the way for more nuanced data analysis methodologies. In conclusion, this research underscores the pivotal role of network platforms in contemporary society, emp

关键词： Network platform tensor-based clustering weight learning multi-linear euclidean