检索结果-南通市图书馆

CAAI Transactions on Intelligence Technology 2024年第4期9卷 960-972页

作者： Jing Li dezheng zhang Yonghong Xie Aziguli Wulamu Yao zhang School of Computer and Communication Engineering University of Science and Technology BeijingBeijingChina Beijing Key Laboratory of Knowledge Engineering for Materials Science University of Science and Technology BeijingBeijingChina University of Alberta EdmontonAlbertaCanada

Sentiment analysis is a fine‐grained analysis task that aims to identify the sentiment polarity of a specified *** methods in Chinese sentiment analysis tasks only consider sentiment features from a single pole and scale and thus cannot fully exploit and utilise sentiment feature information,making their performance less than *** resolve the problem,the authors propose a new method,GP‐FMLNet,that integrates both glyph and phonetic information and design a novel feature matrix learning process for phonetic features with which to model words that have the same pinyin information but different glyph *** method solves the problem of misspelling words influencing sentiment polarity prediction ***,the authors iteratively mine character,glyph,and pinyin features from the input comments ***,the authors use soft attention and matrix compound modules to model the phonetic features,which empowers their model to keep on zeroing in on the dynamic‐setting words in various positions and to dispense with the impacts of the deceptive‐setting ***-periments on six public datasets prove that the proposed model fully utilises the glyph and phonetic information and improves on the performance of existing Chinese senti-ment analysis algorithms.

关键词： aspect‐level sentiment analysis deep learning feature extraction glyph and phonetic feature matrix compound learning

Mn incorporated RuO2 nanocrystals as an efficient and stable bifunctional electrocatalyst for oxygen evolution reaction and hydrogen evolution reaction in acid and alkaline

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 Chemical Letters 2024年第10期35卷 220-226页

作者： Jing Cao dezheng zhang Bianqing Ren Ping Song Weilin Xu State Key Laboratory of Electroanalytical Chemistry Changchun Institute of Applied Chemistry Chinese Academy of Sciences University of Science and Technology of China

The development of efficient and stable bifunctional overall water-splitting is a crucial goal for clean and renewable energy,which is a challenging ***,we report an Mn-incorporated RuO2（MnRuO2） catalyst for highly efficient electrocatalytic oxygen evolution reaction （OER） and hydrogen evolution reaction （HER） in acid and alkaline *** from a more electrochemical active area with the incorporation of Mn,the Mn-RuO2required an overpotential of 200 mV to attain a current density of 10 mA/cm2for OER in *** result indicates that the doping of Mn into RuO2can enhance the OER *** acidic overall water-splitting electrolyzer with good stability constructed by bifunctional Mn-RuO2only requires a cell voltage of 1.50 V to afford 10 m A/cm2and can operate stably for 50 h at50 mA/cm2,which is better than the state-of-the-art Ru-based ***,the Mn-Ru O2exhibits excellent HER and OER activity in alkaline media,and it shows superior activity and durability for overall water-splitting,only needing a cell voltage of 1.49 V to attain 10 m A/*** present work provides an efficient approach to designing and constructing efficient Ru-based electrocatalysts for overall water-splitting.

关键词： Acid and alkaline Hydrogen evolution reaction Overall water splitting Oxygen evolution reaction RuO2-based solid solution electrocatalysts

Heterostructured Pt-Ni_(3)Mo_(3)N formed via ammonia-containing polyoxometalates for highly efficient electrocatalytic hydrogen evolution in acid medium

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Energy Chemistry 2024年第5期92卷 698-704页

作者： Bianqing Ren Xue Gong Jing Cao dezheng zhang Zizhun Wang Ping Song Ce Han Weilin Xu State Key Laboratory of Electroanalytical Chemistry Jilin Province Key Laboratory of Low Carbon Chemical PowerChangchun Institute of Applied ChemistryChinese Academy of SciencesChangchun 130022JilinChina School of Applied Chemistry and Engineering University of Science and Technology of ChinaHefei 230026AnhuiChina Key Laboratory of Automobile Materials MOE School of Materials Science&EngineeringJilin Provincial International Cooperation Key Laboratory of High-Efficiency Clean Energy MaterialsElectron Microscopy Centerand International Center of Future ScienceJilin UniversityChangchun 130012JilinChina

Constructing heterostructured nanohybrid is considered as a prominent route to fabricate alternative electrocatalysts to commercial Pt/C for hydrogen evolution reaction(HER).In this work,(NH_(4))_(4)[NiH_(6)Mo_(6)O_(4)]·5H_(2)O polyoxometalates(NiMo_(6))are adopted as the cluster precursors for simple fabrication of heterostructured Pt-Ni_(3)Mo_(3)N nanohybrids supported by carbon black(Pt-Ni_(3)Mo_(3)N/C)without using additional N *** improved porosity and enhanced electronic interaction of Pt-Ni_(3)Mo_(3)N/C should be attributed to the integration of Pt with NiMo_(6),which favors the mass transport,promotes the formation of exposed catalytic sites,and benefits the regulation of intrinsic ***,the as-obtained Pt-Ni_(3)Mo_(3)N/C exhibits impressive and durable HER performance as indicated by the low overpotential of 13.7 mV at the current density of 10 mA cm^(-2) and the stable overpotential during continuous working at 100 mA cm^(-2) for 100 *** work provides significant insights for the synthesis of new highly active heterostructured electrocatalysts for renewable energy devices.

关键词： Polyoxometalates Cluster precursors Heterostructured nanohybrids Hydrogen evolution reaction Electrocatalysis

Recent progress in single-molecule fluorescence technology in nanocatalysis

在线全文

学校读者我要写书评

暂无评论

Nano Research 2022年第12期15卷 10316-10327页

作者： Jing Cao dezheng zhang Weilin Xu State Key Laboratory of Electroanalytical Chemistry and Jilin Province Key Laboratory of Low Carbon Chemical Power Changchun Institute of Applied ChemistryChinese Academy of SciencesChangchun130022China University of Science and Technology of China Hefei230026China

Nanoparticles(NPs)play a vital role in the energy catalysis process,so understanding the heterogeneous catalytic properties of nanocatalysts is of great significance for rationally guiding the design of ***,the traditional method obtains the average information based on the whole and cannot study the catalytic activity of a single *** is critical to investigate the catalytic activity of individual nanoparticles using in situ *** review summarizes some of ***'s recent accomplishments in studying the catalytic behavior of nanoparticles at the single-particle level using single-molecule fluorescence microscopy(SMFM).These achievements include revealing the effect of size,shape,and surface atoms of Pd nanoparticles on catalytic kinetics and dynamics as well as obtaining the activation energy of single Au nanoparticles for catalytic reactions by single-molecule *** is the first time to study the kinetics and dynamics of single-atom Pt ***,the method was extended to study the Pt deactivation process for hydrogen oxidation reactions as well as the catalytic kinetics of two-electron oxygen reduction reactions of individual Fe3O4 nanoparticles in ***,single-molecule super-resolution techniques were used to observe the evolution of the activity of single Sb doped TiO2 nanorod *** studies are of guiding significance for in-depth understanding and realization of rational design of optimal catalysts.

关键词： single-molecule fluorescence microscopy nanocatalysis kinetics and dynamics activation energy electrocatalysis super-resolution imaging

一种基于FPGA的深度神经网络硬件加速器系统

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

空间控制技术与应用 2024年第2期50卷 83-92页

作者：张雨豪叶有时彭宇张德正阎之泓王东北京交通大学北京100080 北京控制工程研究所北京100094

深度神经网络目标检测算法计算复杂度高、模型复杂,对硬件平台的算力有很高需求,针对以上问题,设计了一种基于现场可编程门阵列(field programmable gate array,FPGA)芯片的硬件专用加速器.通过软硬件协同方法,设计具有高并行度及深度... 详细信息

深度神经网络目标检测算法计算复杂度高、模型复杂,对硬件平台的算力有很高需求,针对以上问题,设计了一种基于现场可编程门阵列(field programmable gate array,FPGA)芯片的硬件专用加速器.通过软硬件协同方法,设计具有高并行度及深度流水的片上架构,并使用模型量化、结构优化等方法对神经网络模型进行优化.在所设计的加速器系统中进行神经网络目标检测算法的部署,实现了高数据吞吐率、低功率消耗的FPGA神经网络计算,且模型精度损失低于1.2%,为在低能耗嵌入式平台上部署深度神经网络目标检测算法提供了有效解决方案,可广泛应用于机载、星载智能计算设备.

关键词： FPGA 神经网络硬件加速器目标检测

连续时间量子行走算法在截断单形晶格上的搜索研究

在线全文

学校读者我要写书评

暂无评论

现代信息科技 2024年第6期8卷 74-77,81页

作者：朱轩民张德政贵州财经大学贵州贵阳550025

为证明连续时间量子行走算法在结构型数据库上的搜索可以实现二次加速的效果,对结构型数据库中的截断单形晶格类型,进行了连续时间量子行走算法的应用研究。首先对截断单形晶格进行对称性分析,确定系统演化所处的希尔伯特空间,然后用哈... 详细信息

为证明连续时间量子行走算法在结构型数据库上的搜索可以实现二次加速的效果,对结构型数据库中的截断单形晶格类型,进行了连续时间量子行走算法的应用研究。首先对截断单形晶格进行对称性分析,确定系统演化所处的希尔伯特空间,然后用哈密顿量本征态与基础态的平方叠加、和简并微扰理论两种方法来求解系统演化需要的临界跳跃率。最后通过对图中的边进行加权的方法,合并了量子搜索的步骤,缩短了系统演化的时间,从而实现了平方加速的效果,并表明了边的权重对量子搜索过程的影响。

关键词：量子计算量子搜索连续时间量子行走算法结构型数据库

基于PSO和滑模观测器的永磁同步电机预测控制

在线全文

学校读者我要写书评

暂无评论

煤矿机械 2024年第5期45卷 192-195页

作者：王伟程卫健张德正中煤科工集团常州研究院有限公司江苏常州213015 天地(常州)自动化股份有限公司江苏常州213015

针对传统永磁同步电机(PMSM)无差拍电流预测控制中参数失配会导致电流误差甚至系统失稳的问题,提出了一种基于粒子群算法(PSO)和滑模观测器的预测控制。基于无差拍电流预测控制算法,建立了电流预测控制系统,并分析了电阻、电感和转子磁... 详细信息

针对传统永磁同步电机(PMSM)无差拍电流预测控制中参数失配会导致电流误差甚至系统失稳的问题,提出了一种基于粒子群算法(PSO)和滑模观测器的预测控制。基于无差拍电流预测控制算法,建立了电流预测控制系统,并分析了电阻、电感和转子磁链参数对dq轴电流误差的影响。通过建立滑模观测器,利用观测到的系统扰动对参考电压进行补偿,提高了系统抗参数扰动的能力。针对观测器参数难以整定问题,提出基于PSO优化观测器参数的方法。最后通过仿真验证了所提方法的可行性。

关键词：无差拍预测控制 PMSM 滑模观测器 PSO 参数失配

Classification with ensembles and case study on functional magnetic resonance imaging

在线全文

学校读者我要写书评

暂无评论

Digital Communications and Networks 2022年第1期8卷 80-86页

作者： Adnan OM.Abuassba zhang dezheng Hazrat Ali Fan zhang Khan Ali Department of Computer Science Arab Open University-PalestineRamallahPalestine Department of Computer Science and Communication Engineering University of Science and Technology BeijingChina and Beijing Key Laboratory of Knowledge Engineering for Materials ScienceBeijing100083China Department of Electrical and Computer Engineering COMSATS University IslamabadAbbottabad CampusAbbottabadPakistan

The ensemble is a technique that strategically combines basic models to achieve better accuracy ***,combination methods,and selection topology are the main factors determining ensemble ***,it is a challenging task to design an efficient ensemble *** though numerous paradigms have been proposed to classify ensemble schemes,there is still much room for *** paper proposes a general framework for creating ensembles in the context of ***,the ensemble framework consists of four stages:objectives,data preparing,model training,and model *** is comprehensive to design diverse *** proposed ensemble approach can be used for a wide variety of machine learning *** validate our approach on real-world *** experimental results show the efficiency of the proposed approach.

关键词： Classification Ensemble learning Extreme learning machine

A Heterogeneous Ensemble of Extreme Learning Machines with Correntropy and Negative Correlation

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 2017年第6期22卷 691-701页

作者： Adnan O.M.Abuassba Yao zhang Xiong Luo dezheng zhang Wulamu Aziguli School of Computer and Communication Engineering University of Science and Technology Beijing(USTB)Beijing 100083China Beijing Key Laboratory of Knowledge Engineering for Materials Science Beijing 100083China School of Engineering New York UniversityBrooklynNY 11201USA

The Extreme Learning Machine（ELM） is an effective learning algorithm for a Single-Layer Feedforward Network（SLFN）. It performs well in managing some problems due to its fast learning speed. However, in practical applications, its performance might be affected by the noise in the training data. To tackle the noise issue, we propose a novel heterogeneous ensemble of ELMs in this article. Specifically, the correntropy is used to achieve insensitive performance to outliers, while implementing Negative Correlation Learning（NCL） to enhance diversity among the ensemble. The proposed Heterogeneous Ensemble of ELMs（HE2 LM） for classification has different ELM algorithms including the Regularized ELM（RELM）, the Kernel ELM（KELM）, and the L2-norm-optimized ELM（ELML2）. The ensemble is constructed by training a randomly selected ELM classifier on a subset of the training data selected through random resampling. Then, the class label of unseen data is predicted using a maximum weighted sum approach. After splitting the training data into subsets, the proposed HE2 LM is tested through classification and regression tasks on real-world benchmark datasets and synthetic datasets. Hence, the simulation results show that compared with other algorithms, our proposed method can achieve higher prediction accuracy, better generalization, and less sensitivity to outliers.

关键词： Extreme Learning Machine(ELM) ensemble classification correntropy negative correlation

Auxiliary Diagnosis Based on the Knowledge Graph of TCM Syndrome

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2020年第10期65卷 481-494页

作者： Yonghong Xie Liangyuan Hu Xingxing Chen Jim Feng dezheng zhang School of Computer and Communication Engineering University of Science and Technology BeijingBeijing100083China School of Automation and Electrical Engineering University of Science and Technology BeijingBeijing100083China Beijing Key Laboratory of Knowledge Engineering for Materials Science Beijing100083China Amphenol Global Interconnect Systems San JoseCA 95131USA

As one of the most valuable assets in China,traditional medicine has a long history and contains pieces of *** diagnosis and treatment of Traditional Chinese Medicine(TCM)has benefited from the natural language processing *** paper proposes a knowledge-based syndrome reasoning method in computer-assisted *** method is based on the established knowledge graph of TCM and this paper introduces the reinforcement learning algorithm to mine the hidden relationship among the entities and obtain the reasoning *** to this reasoning path,we could infer the path from the symptoms to the syndrome and get all possibilities via the relationship between symptoms and ***,this study applies the Term Frequency-Inverse Document Frequency(TF-IDF)idea to the computer-assisted diagnosis of TCM for the score of syndrome ***,combined with symptoms,syndrome,and causes,the disease could be confirmed comprehensively by voting,and the experiment shows that the system can help doctors and families to disease diagnosis effectively.

关键词： Knowledge graph reinforcement learning auxiliary diagnosis inference path