检索结果-南通市图书馆

Improve Chinese Aspect Sentiment Quadruplet Prediction via Instruction Learning Based on Large Generate Models

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 3391-3412页

作者： Zhaoliang Wu Yuewei Wu Xiaoli Feng Jiajun Zou Fulian Yin College of Information and communication Engineering Communication University of ChinaBeijing100024China the state key laboratory of media convergence and communication Communication University of ChinaBeijing100024China Department of Electronic Engineering Tsinghua UniversityBeijing100084China

Aspect-Based Sentiment Analysis(ABSA)is a fundamental area of research in Natural Language Processing(NLP).Within ABSA,Aspect Sentiment Quad Prediction(ASQP)aims to accurately identify sentiment quadruplets in target sentences,including aspect terms,aspect categories,corresponding opinion terms,and sentiment ***,most existing research has focused on English ***,while ASQP has seen significant progress in English,the Chinese ASQP task has remained relatively *** inspiration from methods applied to English ASQP,we propose Chinese generation templates and employ prompt-based instruction learning to enhance the model’s understanding of the task,ultimately improving ASQP performance in the Chinese ***,under the same pre-training model configuration,our approach achieved a 5.79%improvement in the F1 score compared to the previously leading ***,when utilizing a larger model with reduced training parameters,the F1 score demonstrated an 8.14%***,we suggest a novel evaluation metric based on the characteristics of generative models,better-reflecting model *** results validate the effectiveness of our approach.

关键词： ABSA ASQP LLMs sentiment analysis Chinese comments

Deep Video Harmonization by Improving Spatial-temporal Consistency

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Machine Intelligence Research 2024年第1期21卷 46-54页

作者： Xiuwen Chen Li Fang Long Ye Qin Zhang key laboratory of media Audio and Video Ministry of Education Communication University of ChinaBeijing 100024China

Video harmonization is an important step in video editing to achieve visual consistency by adjusting foreground appear-ances in both spatial and temporal *** methods always only harmonize on a single scale or ignore the inaccuracy of flow estimation,which leads to limited harmonization *** this work,we propose a novel architecture for video harmoniza-tion by making full use of spatiotemporal features and yield temporally consistent harmonized *** introduce multiscale harmon-ization by using nonlocal similarity on each scale to make the foreground more consistent with the *** also propose a fore-ground temporal aggregator to dynamically aggregate neighboring frames at the feature level to alleviate the effect of inaccurate estim-ated flow and ensure temporal *** experimental results demonstrate the superiority of our method over other state-of-the-art methods in both quantitative and visual comparisons.

关键词： Harmonization temporal consistency video editing video composition nonlocal similarity

Aspect-Level Sentiment Analysis Based on Deep Learning

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 3743-3762页

作者： Mengqi Zhang Jiazhao Chai Jianxiang Cao Jialing Ji Tong Yi Department of communication University of New South WalesSydney2052Australia Higher Vocational Education Publishing Division Higher Education PressBeijing100029China state key laboratory of media convergence and communication Communication University of ChinaBeijing100020China key Lab of Education Blockchain and Intelligent Technology Ministry of EducationGuangxi Normal UniversityGuilin541004China

In recent years,deep learning methods have developed rapidly and found application in many fields,including natural language *** the field of aspect-level sentiment analysis,deep learning methods can also greatly improve the performance of ***,previous studies did not take into account the relationship between user feature extraction and contextual *** address this issue,we use data feature extraction and deep learning combined to develop an aspect-level sentiment analysis *** be specific,we design user comment feature extraction(UCFE)to distill salient features from users’historical comments and transform them into representative user feature ***,the aspect-sentence graph convolutional neural network(ASGCN)is used to incorporate innovative techniques for calculating adjacency matrices;meanwhile,ASGCN emphasizes capturing nuanced semantics within relationships among aspect words and syntactic dependency ***,three embedding methods are devised to embed the user feature vector into the ASGCN *** empirical validations verify the effectiveness of these models,consistently surpassing conventional benchmarks and reaffirming the indispensable role of deep learning in advancing sentiment analysis methodologies.

关键词： Aspect-level sentiment analysis deep learning graph convolutional neural network user features syntactic dependency tree

Fine-Grained Emotion Prediction for Movie and Television scene images

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

The Journal of China Universities of Posts and Telecommunications 2024年第3期31卷 43-55页

作者： Su Zhibin Zhou Xuanye Liu Bing Ren Hui state key laboratory of media convergence and communication Communication University of ChinaBeijing 100024China key laboratory of Acoustic Visual Technology and Intelligent Control System Communication University of ChinaBeijing 100024China School of Information and communication Engineering Communication University of ChinaBeijing 100024China

For the task of content retrieval,analysis and generation of film and television scene images in the field of intelligent editing,fine-grained emotion recognition and prediction of images is of great *** this paper,the fusion of traditional perceptual features,art features and multi-channel deep learning features are used to reflect the emotion expression of different levels of the *** addition,the integrated learning model with stacking architecture based on linear regression coefficient and sentiment correlations,which is called the LS-stacking model,is proposed according to the factor association between multi-dimensional *** experimental results prove that the mixed feature and LS-stacking model can predict well on the 16 emotion categories of the self-built image *** study improves the fine-grained recognition ability of image emotion by computers,which helps to increase the intelligence and automation degree of visual retrieval and post-production system.

关键词： fine-grained emotion prediction movie and television scene images stacking model linear regression

同方期刊数据库

Enhancing Deep Learning Semantics:The Diffusion Sampling and Label-Driven Co-Attention Approach

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第5期79卷 1939-1956页

作者： ChunhuaWang Wenqian Shang Tong Yi Haibin Zhu state key laboratory of media convergence and communication Communication University of ChinaBeijing100024China School of Computer and Cyber Sciences Communication University of ChinaBeijing100024China School of Computer Science and Engineering Guangxi Normal UniversityGuilin541004China Department of Computer Science Nipissing UniversityNorth BayON P1B 8L7Canada

The advent of self-attention mechanisms within Transformer models has significantly propelled the advancement of deep learning algorithms,yielding outstanding achievements across diverse ***,self-attention mechanisms falter when applied to datasets with intricate semantic content and extensive dependency *** response,this paper introduces a Diffusion Sampling and Label-Driven Co-attention Neural Network(DSLD),which adopts a diffusion sampling method to capture more comprehensive semantic information of the ***,themodel leverages the joint correlation information of labels and data to introduce the computation of text representation,correcting semantic representationbiases in thedata,andincreasing the accuracyof semantic ***,the model computes the corresponding classification results by synthesizing these rich data semantic *** on seven benchmark datasets show that our proposed model achieves competitive results compared to state-of-the-art methods.

关键词： Semantic representation sampling attention label-driven co-attention attention mechanisms

All-dielectric terahertz metasurface governed by bound states in the continuum with high-Q factor

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

The Journal of China Universities of Posts and Telecommunications 2024年第2期31卷 44-54页

作者： Li Kun Zhao Xu Wang Meng Jia Xingning Zhao Ruiqi Lu Guizhen School of Integrated Circuit Science and Engineering Tianjin University of TechnologyTianjin 300384China School of Electronic and Electrical Engineering Ningxia UniversityYinchuan 750021China QILU Aerospace Information Research Institute Chinese Academy of Sciences(CAS)Jinan 250101China state key laboratory of media convergence and communication Communication University of ChinaBeijing 100024China

The method of terahertz(THz)resonance with a high-quality(high-Q)factor offers a vital physical mechanism for metasurface sensors and other high-Q factor ***,it is challenging to excite the resonance with a high-Q factor in metasurfaces with proper sensitivity as well as figure of merit(FOM)***,an all-dielectric metasurface composed of two asymmetrical rectangular blocks is *** and silicon are the materials applied for the substrate and cuboids *** distinct resonance governed by bound states in the continuum(BIC)is excited by forming an asymmetric cluster by a novel hybrid method of cutting and moving the *** investigation focuses on analyzing the transmission spectra of the metasurface under different variations in structural parameters and the loss of silicon refractive *** the proposed defective metasurface serves as a transmittance sensor,it shows a Q factor of 1.08×10^(4)and achieves an FOM up to 4.8×10^(6),which is obtained under the asymmetric parameter equalling 1μ***,the proposed defective metasurface is sensitive to small changes in refractive *** the thickness of the analyte is 180μm,the sensitivity reaches a maximum value of 578 GHz/***,the proposed defective metasurface exhibits an extensive number of possible applications in the filters,biomedical diagnosis,security screening,and so on.

关键词： terahertz(THz)metasurface high-quality(high-Q)factor bound states in the continuum figure of merit(FOM)

同方期刊数据库

Building mm Wave on the Evolving C-V2X:mm Wave NR-V2X

在线全文

学校读者我要写书评

暂无评论

China communications 2024年第1期21卷 88-101页

作者： Shanzhi Chen Xinxin He Rui Zhao Jinling Hu Xin Zhang China Information and communication Technology Group Co. Ltd.(CICT)the State Key Laboratory of Wireless Mobile CommunicationsBeijing 100083China Beijing laboratory of Advanced Information Networks Beijing Key Laboratory of Network System Architecture and ConvergenceBeijing University of Posts and TelecommunicationsBeijing 100876China CICT Connected and Intelligent Technologies Co. LtdChina Academy of Telecommunication TechnologyBeijing 100191China CICT Mobile communications Technology Co.Ltd.

Since the evolving standardization of cellular V2X(C-V2X)technologies is continuously developed by 3GPP,research on new radio(NR-V2X)has been on schedule by academic and *** millimeter wave(mm Wave)frequency band is supposed to provide large transmission bandwidth,yet the development of mm Wave band in NR-V2X is still in preliminary *** this article,a comprehensive discussion of mm Wave NR-V2X is given covering trends,standardization landscape,and enabling technologies,aiming at tackling the challenge of channel modeling,directional channel access,beamforming and blockage *** vision of mm Wave NR-V2X is to fully support the development of automated driving,holographic control display and improved in-car infotainment for the future.

关键词： C-V2X enabling technologies mmWave NR-V2X

Diffraction deep neural network-based classification for vector vortex beams

在线全文

学校读者我要写书评

暂无评论

Chinese Physics B 2024年第3期33卷 387-392页

作者：彭怡翔陈兵王乐赵生妹 Institute of Signal Processing and Transmission Nanjing University of Posts and Telecommunications (NJUPT)Nanjing 210003China key laboratory of Broadband Wireless communication and Sensor Network Technology Ministry of EducationNanjing 210003China National laboratory of Solid state Microstructures Nanjing UniversityNanjing 210093China

The vector vortex beam(VVB)has attracted significant attention due to its intrinsic diversity of information and has found great applications in both classical and quantum ***,a VVB is unavoidably affected by atmospheric turbulence(AT)when it propagates through the free-space optical communication environment,which results in detection errors at the *** this paper,we propose a VVB classification scheme to detect VVBs with continuously changing polarization states under AT,where a diffractive deep neural network(DDNN)is designed and trained to classify the intensity distribution of the input distorted VVBs,and the horizontal direction of polarization of the input distorted beam is adopted as the feature for the classification through the *** numerical simulations and experimental results demonstrate that the proposed scheme has high accuracy in classification *** energy distribution percentage remains above 95%from weak to medium AT,and the classification accuracy can remain above 95%for various strengths of *** has a faster convergence and better accuracy than that based on a convolutional neural network.

关键词： vector vortex beam diffractive deep neural network classification atmospheric turbulence

Cooperative User-Scheduling and Resource Allocation Optimization for Intelligent Reflecting Surface Enhanced LEO Satellite communication

在线全文

学校读者我要写书评

暂无评论

China communications 2024年第2期21卷 227-244页

作者： Meng Meng Bo Hu Shanzhi Chen Jianyin Zhang state key Lab of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China state key Lab of Wireless Mobile communication China Academy of Telecommunication TechnologyBeijing 100191China Research Institute of China Mobile Beijing 100053China

Lower Earth Orbit(LEO) satellite becomes an important part of complementing terrestrial communication due to its lower orbital altitude and smaller propagation delay than Geostationary satellite. However, the LEO satellite communication system cannot meet the requirements of users when the satellite-terrestrial link is blocked by obstacles. To solve this problem, we introduce Intelligent reflect surface(IRS) for improving the achievable rate of terrestrial users in LEO satellite communication. We investigated joint IRS scheduling, user scheduling, power and bandwidth allocation(JIRPB) optimization algorithm for improving LEO satellite system *** optimization problem of joint user scheduling and resource allocation is formulated as a non-convex optimization problem. To cope with this problem, the nonconvex optimization problem is divided into resource allocation optimization sub-problem and scheduling optimization sub-problem firstly. Second, we optimize the resource allocation sub-problem via alternating direction multiplier method(ADMM) and scheduling sub-problem via Lagrangian dual method ***, we prove that the proposed resource allocation algorithm based ADMM approaches sublinear convergence theoretically. Finally, we demonstrate that the proposed JIRPB optimization algorithm improves the LEO satellite communication system throughput.

关键词： convex optimization intelligent reflecting surface LEO satellite communication OFDM