检索结果-南通市图书馆

Journal of Thermal Science 2018年第2期27卷 135-145页

作者： wang yufeng CAI Le wang Songtao ZHOU Xun Engine Aerodynamics Research Centre Harbin Institute of Technology Harbin 150001 China

Unsteady numerical simulations of a high-load transonic turbine stage have been carried out to study the influences of vane trailing edge outer-extending shockwave on rotor blade leading edge film cooling performance. The turbine stage used in this paper is composed of a vane section and a rotor one which are both near the root section of a transonic high-load turbine stage. The Mach number is 0.94 at vane outlet, and the relative Mach number is above 1.10 at rotor outlet. Various positions and oblique angles of film cooling holes were investigated in this research. Results show that the cooling efficiency on the blade surface of rotor near leading edge is significantly affected by vane trailing edge outer-extending shockwave in some cases. In the cases that film holes are close to leading edge, cooling performance suffers more from the sweeping vane trailing edge outer-extending *** addition, coolant flow ejected from oblique film holes is harder to separate from the blade surface of rotor, and can cover more blade area even under the effects of sweeping vane trailing edge shockwave. As a result, oblique film holes can provide better film cooling performance than vertical film holes do near the leading edge on turbine blade which is swept by shockwaves.

关键词： transonic turbine trailing edge shock wave ftlm cooling unsteady simulation

CAN:Effective Cross Features by Global Attention Mechanism and Neural Network for Ad Click Prediction

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2022年第1期27卷 186-195页

作者： Wenjie Cai yufeng wang Jianhua Ma Qun Jin Nanjing University of Posts and Telecommunications(NJUPT) Nanjing210003China Digital Media Department in the Faculty of Computer and Information Sciences Hosei UniversityTokyo 163-8001Japan Networked Information Systems Laboratory Department of Human Informatics and Cognitive SciencesWaseda UniversityTokyo 163-8001Japan

Online advertising click-through rate(CTR) prediction is aimed at predicting the probability of a user clicking an ad,and it has undergone considerable development in recent *** of the hot topics in this area is the construction of feature interactions to facilitate accurate *** machine provides second-order feature interactions by linearly multiplying hidden feature ***,real-world data present a complex and nonlinear ***,second-order feature interactions are unable to represent cross information *** drawback has been addressed using deep neural networks(DNNs),which enable high-order nonlinear feature ***,DNN-based feature interactions cannot easily optimize deep structures because of the absence of cross information in the original *** this study,we propose an effective CTR prediction algorithm called CAN,which explicitly exploits the benefits of attention mechanisms and DNN *** attention mechanism is used to provide rich and expressive low-order feature interactions and facilitate the optimization of DNN-based predictors that implicitly incorporate high-order nonlinear feature *** experiments using two real datasets demonstrate that our proposed CAN model performs better than other cross feature-and DNN-based predictors.

关键词： click-through rate prediction global attention mechanism feature interaction neural network

Personalized Real-Time Movie Recommendation System:Practical Prototype and Evaluation

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2020年第2期25卷 180-191页

作者： Jiang Zhang yufeng wang Zhiyuan Yuan Qun Jin Nanjing University of Posts and Telecommunications(NUPT) Nanjing 210003China Networked Information Systems Laboratory Department of Human Informatics and Cognitive SciencesFaculty of Human SciencesWaseda UniversityTokyo 163-8001Japan

With the eruption of big data,practical recommendation schemes are now very important in various fields,including e-commerce,social networks,and a number of web-based ***,there exist many personalized movie recommendation schemes utilizing publicly available movie datasets(e.g.,MovieLens and Netflix),and returning improved performance metrics(e.g.,Root-Mean-Square Error(RMSE)).However,two fundamental issues faced by movie recommendation systems are still neglected:first,scalability,and second,practical usage feedback and verification based on real *** particular,Collaborative Filtering(CF)is one of the major prevailing techniques for implementing recommendation ***,traditional CF schemes suffer from a time complexity problem,which makes them bad candidates for real-world recommendation *** this paper,we address these two ***,a simple but high-efficient recommendation algorithm is proposed,which exploits users1 profile attributes to partition them into several *** each cluster,a virtual opinion leader is conceived to represent the whole cluster,such that the dimension of the original useritem matrix can be significantly reduced,then a Weighted Slope One-VU method is designed and applied to the virtual opinion leader-item matrix to obtain the recommendation *** to traditional clusteringbased CF recommendation schemes,our method can significantly reduce the time complexity,while achieving comparable recommendation ***,we have constructed a real personalized web-based movie recommendation system,MovieWatch,opened it to the public,collected user feedback on recommendations,and evaluated the feasibility and accuracy of our system based on this real-world data.

关键词： movie recommendation system collaborative filtering real-time virtual opinion leader data mining

SGNR: A Social Graph Neural Network Based Interactive Recommendation Scheme for E-Commerce

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2023年第4期28卷 786-798页

作者： Dehua Ma yufeng wang Jianhua Ma Qun Jin School of Communications and lnformation Engineering Nanjing University of Posts and TelecommunicationsNanjing 210003China Department the Faculty of Computer and Information SciencesHosei UniversityTokyo 194-0298Japan Networked Information Systems Laboratory Department of Human Informatics and Cognitive SciencesFaculty of Human SciencesWaseda UniversityTokyo 169-8050Japan

Interactive Recommendation(IR)formulates the recommendation as a multi-step decision-making process which can actively utilize the individuals’feedback in multiple steps and optimize the long-term user benefit of *** Reinforcement Learning(DRL)has witnessed great application in IR for ***,user cold-start problem impairs the learning process of the DRL-based recommendation ***,most existing DRL-based recommendations ignore user relationships or only consider the single-hop social relationships,which cannot fully utilize the social *** fact that those schemes can not capture the multiple-hop social relationships among users in IR will result in a sub-optimal *** address the above issues,this paper proposes a Social Graph Neural network-based interactive Recommendation scheme(SGNR),which is a multiple-hop social relationships enhanced DRL *** this framework,the multiple-hop social relationships among users are extracted from the social network via the graph neural network which can sufficiently take advantage of the social network to provide more personalized recommendations and effectively alleviate the user cold-start *** experimental results on two real-world datasets demonstrate that the proposed SGNR outperforms other state-of-the-art DRL-based methods that fail to consider social relationships or only consider single-hop social relationships.

关键词： Interactive Recommendation(IR) Deep Reinforcement Learning(DRL) Graph Neural Network(GNN)

An Integrated Incentive Framework for Mobile Crowdsourced Sensing

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2016年第2期21卷 146-156页

作者： Wei Dai yufeng wang Qun Jin Jianhua Ma College of Telecommunications and Information Engineering Nanjing University of Posts and TelecommunicationsNanjing 210003China. Department of Human Informatics and Cognitive Sciences Waseda UniversitySaitama 359-1192Japan. Faculty of Computer&Information Sciences Hosei UniversityTokyo 184-8584Japan

Currently, mobile devices （e.g., smartphones） are equipped with multiple wireless interfaces and rich builtin functional sensors that possess powerful computation and communication capabilities, and enable numerous Mobile Crowdsourced Sensing （MCS） applications. Generally, an MCS system is composed of three components： a publisher of sensing tasks, crowd participants who complete the crowdsourced tasks for some kinds of rewards, and the crowdsourcing platform that facilitates the interaction between publishers and crowd participants. Incentives are a fundamental issue in MCS. This paper proposes an integrated incentive framework for MCS, which appropriately utilizes three widely used incentive methods： reverse auction, gamification, and reputation updating. Firstly, a reverse-auction-based two-round participant selection mechanism is proposed to incentivize crowds to actively participate and provide high-quality sensing data. Secondly, in order to avoid untruthful publisher feedback about sensing-data quality, a gamification-based verification mechanism is designed to evaluate the truthfulness of the publisher＇s feedback. Finally, the platform updates the reputation of both participants and publishers based on their corresponding behaviors. This integrated incentive mechanism can motivate participants to provide high-quality sensed contents, stimulate publishers to give truthful feedback, and make the platform profitable.

关键词： mobile crowdsourced sensing incentive mechanism reverse auction gamification reputation updating

Inclusion engineering and metallurgy of calcium treatment

在线全文

学校读者我要写书评

暂无评论

Journal of Iron and Steel Research International 2018年第2期25卷 133-145页

作者： Sunday Abraham Rick Bodnar Justin Raines yufeng wang SSAB Americas Muscatine IA 52761 USA

Although a great deal of research has been done in the area of inclusion control during steelmaking, a complete description of the process of inclusion modification and shape control cannot be found in literatures. As a result, the terms ＂inclusion modification＂ and ＂shape control＂ are often used loosely in the technical community. To utilize the full potential offered by a calcium treatment, it is important to understand why an inclusion whose composition has been modified may not necessarily have its shape modified. With the purpose of decoupling the process of inclusion phase change from inclusion shape control, and to provide a complete description of the metallurgy of calcium treatment, a detailed literature survey has been performed on the subject. By coupling the survey with a new concept developed to tailor the cleanliness of steels to the intended applications, the hope is to provide the reader with valuable information on calcium treatment for inclusion modification and shape control.

关键词： Nonmetallic inclusion Cleanliness Calcium treatment Inclusion modification Inclusion shape control

Effect of copper on the development, protein and esterase isozymes of Drosophila melanogaster

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Integrative Zoology 2006年第2期1卷 73-77页

作者： Ling DING yufeng wang College of Life Science Central China Normal UniversityWuhanChina

Copper is an important ingredient in fungicides,which are used for many economic ***,there is concern about the side-effects of copper-based fungicides due to their potential to affect beneficial *** purpose of the present paper was to investigate the effects of copper on the development of the fruit fly,Drosophila *** was found that higher doses of copper significantly prolonged the developmental time of the fruit flies,especially during the larval *** 320 mg L-1 copper-contaminated food was used,most of the larvae died when they were small and before the pupal *** protein contents and esterase isozymes extracted from the larvae changed according to the doses of *** small-molecular-weight protein bands gradually became weaker or were lost as the copper levels ***,low doses of copper stimulated a stronger expression of a few *** results indicate that low doses of copper generally have no lethal effects on *** because a specific group of genes,which encode specific proteins,are probably activated in order to withstand the onslaught of stressful *** high doses of copper in food(e.g.320 mg L-1),fly development and viability are significantly affected.

关键词： copper Drosophila melanogaster esterase isozymes larval development protein

Hypoxic tumor microenvironment:Destroyer of natural killer cell function

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Cancer Research 2024年第2期36卷 138-150页

作者： Yongfei Zhang Feifei Guo yufeng wang Cancer Center the First Hospital of Jilin UniversityChangchun 130021China

In recent years, immunotherapy has made remarkable progress in treating certain tumors and hematological malignancies. However, the efficacy of natural killer(NK) cells, which are an important subset of innate lymphocytes used in anticancer immunotherapy, remains limited. Hypoxia, a critical characteristic of the tumor microenvironment(TME), is involved in tumor development and resistance to radiotherapy, chemotherapy, and immunotherapy. Moreover, hypoxia contributes to the impairment of NK cell function and may be a significant factor that limits their therapeutic effects. Targeted hypoxia therapy has emerged as a promising research area for enhancing the efficacy of NK cell therapy. Therefore, understanding how the hypoxic TME influences NK cell function is crucial for improving antitumor treatment outcomes.

关键词： Hypoxia immunotherapy NK cell tumor microenvironment

Dissymmetric flow phenomenon in a multistrand tundish

在线全文

学校读者我要写书评

暂无评论

Journal of University of Science and Technology Beijing 2007年第6期14卷 490-494页

作者： Mingmei Zhu Guanghua Wen Ping Tang yufeng wang College of Material Science and Engineering Chongqing University Chongqing 400044 China

The dissymmetric flow phenomenon exists in a symmetric multistrand tundish. It was studied by the physical simulation experiment. The fundamental flow characteristic of dissymmetry was analyzed. The asymmetry of the flow field, the temperature field, and the inclusions distribution without flow-control devices （FCDs） were compared with those with FCDs. It is proved that the asymmetry of the flow and temperature field along the outlets at the long range is more obvious. The symmetric FCDs installation has a slight effect on the dissymmetric temperature field, simultaneously, the symmetry of the average residence time and the fluid flow pattern has improved, and the fluid flow in the tundish has been more reasonable. In case of a symmetric multistrand tundish having a large volume, the influence of the dissymmetric phenomenon should be considered and the flow behaviors in the whole tundish should be studied completely.

关键词： multistrand tundish dissymmetry flow phenomenon physical simulation flow-control devices

PI-VEGAN:Physics Informed Variational Embedding Generative Adversarial Networks for Stochastic Differential Equations

在线全文

学校读者我要写书评

暂无评论

Numerical Mathematics(Theory,Methods and Applications) 2023年第4期16卷 931-953页

作者： Ruisong Gao yufeng wang Min Yang Chuanjun Chen School of Mathematics and Information Sciences Yantai UniversityYantaiChina

We present a new category of physics-informed neural networks called physics informed variational embedding generative adversarial network(PI-VEGAN),that effectively tackles the forward,inverse,and mixed problems of stochastic differential *** these scenarios,the governing equations are known,but only a limited number of sensor measurements of the system parameters are *** integrate the governing physical laws into PI-VEGAN with automatic differentiation,while introducing a variational encoder for approximating the latent variables of the actual distribution of the *** latent variables are integrated into the generator to facilitate accurate learning of the characteristics of the stochastic partial *** model consists of three components,namely the encoder,generator,and discriminator,each of which is updated alternatively employing the stochastic gradient descent *** evaluate the effectiveness of PI-VEGAN in addressing forward,inverse,and mixed problems that require the concurrent calculation of system parameters and *** results demonstrate that the proposed method achieves satisfactory stability and accuracy in comparison with the previous physics-informed generative adversarial network(PI-WGAN).

关键词： Stochastic differential equations physics-informed variational inference generative adversarial networks inverse problems