检索结果-内蒙古大学图书馆

Bayesian estimation-based sentiment word embedding model for sentiment analysis

CAAI Transactions on Intelligence Technology 2022年第2期7卷 144-155页

作者： Jingyao Tang Yun Xue Ziwen Wang Shaoyang Hu Tao Gong Yinong Chen Haoliang Zhao Luwei Xiao Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials School of Physics and Telecommunication EngineeringSouth China Normal UniversityGuangzhouChina College of Mathematics and Informatics&College of Software Engineering South China Agricultural UniversityGuangzhouChina School of Foreign Languages Zhejiang University of Finance&EconomicsHangzhouZhejiangChina Educational Testing Service PrincetonNew JerseyUSA School of Computing Informatics and Decision Systems EngineeringArizona State UniversityTempeUSA

Sentiment word embedding has been extensively studied and used in sentiment analysis ***,most existing models have failed to differentiate high-frequency and low-frequency ***,the sentiment information of low-frequency words is insufficiently captured,thus resulting in inaccurate sentiment word embedding and degradation of overall performance of sentiment analysis.A Bayesian estimation-based sentiment word embedding(BESWE)model,which aims to precisely extract the sentiment information of low-frequency words,has been *** the model,a Bayesian estimator is constructed based on the co-occurrence probabilities and sentiment proba-bilities of words,and a novel loss function is defined for sentiment word embedding *** experimental results based on the sentiment lexicons and Movie Review dataset show that BESWE outperforms many state-of-the-art methods,for example,C&W,CBOW,GloVe,SE-HyRank and DLJT1,in sentiment analysis tasks,which demonstrate that Bayesian estimation can effectively capture the sentiment information of low-frequency words and integrate the sentiment information into the word embedding through the loss *** addition,replacing the embedding of low-frequency words in the state-of-the-art methods with BESWE can significantly improve the performance of those methods in sentiment analysis tasks.

关键词： function embedding estimation

来源：评论

学校读者我要写书评

暂无评论

FreqSense: Universal and Low-Latency Adversarial Example Detection for Speaker Recognition with Interpretability in Frequency Domain

FreqSense: Universal and Low-Latency Adversarial Example Det...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yihuan Huang Yuanzhe Li Yanzhen Ren Weiping Tu Yuhong Yang Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of Education School of Cyber Science and Engineering Wuhan University National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Speaker recognition (SR) systems are particularly vulnerable to adversarial example (AE) attacks. To mitigate these attacks, AE detection systems are typically integrated into SR systems. To overcome the limitations of low detection accuracy, poor generalization, and high latency in existing schemes, this paper proposes FreqSense, an AE detection scheme based on frequency distribution features. FreqSense detects a variety of unknown AE attacks with low latency, and provides interpretability in its detection process. The basic idea of FreqSense is that AE typically introduce carefully designed noise in specific frequency bands that are associated with highly distinctive speaker identities. Therefore, leveraging the distributional variations in these frequency bands can effectively distinguish between AE and benign audio. FreqSense models frequency distribution features by integrating time-frequency transformation technology with a self-attention mechanism and employs a neural network-based classifier to distinguish between AE and benign audio. Experimental results show that FreqSense achieves an overall detection accuracy of 99.2%, surpassing state-of-the-art (SOTA) schemes by 27.2%. When confronting unknown AE attacks, FreqSense achieves a detection accuracy of 98.3% with a latency of just 0.0014 seconds.

关键词： Time-frequency analysis Accuracy Noise Speech recognition Feature extraction Acoustics Speaker recognition Low latency communication Speech processing

来源：评论

学校读者我要写书评

暂无评论

Creating an Inclusive, Vibrant Learning Environment within a Large, software engineering Program - Experiential Learning Experiences Created for Students, Faculty, and Senior Design Coaches & Sponsors

Creating an Inclusive, Vibrant Learning Environment within a...

引用

CoNECD 2021

作者： Bailey, Margaret B. Sharma, Naveen Verschage, Lana Conlon, Timothy T. Department of Mechanical Engineering Kate Gleason College of Engineering Rochester Institute of Technology Department of Software Engineering Golisano College of Computing and Information Sciences Rochester Institute of Technology Xerox Corporation Women in Computing Program Golisano College of Computing and Information Sciences Rochester Institute of Technology

Creating an Inclusive, Vibrant Learning Environment within a Large, software engineering Program - Experiential Learning Experiences Created for Students, Faculty, and Senior Design Coaches & Sponsors This presentation describes several initiatives currently underway within a large software engineering (SE) department at a private university within the northeast US to positively affect levels of inclusiveness within the department's learning environment. Information gathered from faculty, staff, student interviews and focus group discussions, as well as a department goal to improve female freshmen retention in the SE major motivated the SE department chair, departmental academic advisors and faculty, and the college's women in computing director to launch this set of aligned activities in collaboration with the NSF ADVANCE funded program (NSF #1209115) at the university level. The resulting activities ideally enhance diversity and inclusion for students from all underrepresented groups in the program with a focus on groups based on gender. Four experiential learning experiences have been created and concurrently implemented within the SE department to promote an inclusive academic environment. These include: (1) faculty targeted discussions and summer readings, (2) an interactive workshop designed for all first year SE students, (3) a workshop created for the coaches and sponsors, most of who are not regular RIT employees, who directly mentor student teams for the two‐semester senior project course, and (4) the development of a resource flowchart which supports students, faculty, and staff in maintaining an inclusive learning environment within the department. The approach used in creating each is adaptive and the four resulting products are multi‐faceted in regards to target audience, modality of learning experience, and composition of creation team. Furthermore, we will discuss key metrics aligned with the department's goal to measure impacts resulting from these

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Decision Management: Multi-Modal Analytics in a Health Case Study 19

Data-Driven Decision Management: Multi-Modal Analytics in a ...

引用

19th International Conference on Ubiquitous Information Management and Communication, IMCOM 2025

作者： Agomuo, Okechukwu Clement Khan, Zohaib Otuomasirichi, Agomuo Ijeoma Uzoma, Agomuo Kingsley Muzamal, Junaid Hussain Department of Operations and Information Systems University of Utah Salt Lake City United States Dept. of Software Engineering FAST NUCES Lahore Pakistan Department of Agricultural Economics University of Nigeria Enugu Nigeria Center for Environmental Management and Green Energy University of Nigeria Enugu Nigeria Fast School of Computing Fast NUCES Lahore Pakistan

ISBN: (纸本)9798331507817

In an era where data is abundant yet underutilized, organizations across various sectors face challenges in extracting actionable insights from complex, multi-modal datasets. To address these challenges, this study proposes a versatile data analytics framework designed to enhance predictive accuracy and support informed decision-making. The framework integrates structured and unstructured data, leveraging advanced Transformer architectures to capture temporal dynamics and complex interactions within the data. In a healthcare use case, the framework demonstrated its effectiveness by achieving an Accuracy of 0.854, an F1-Score of 0.895, and an AUC-ROC of 0.930, significantly outperforming traditional models. By incorporating SHAP for interpretability, the framework ensures that its predictions are both accurate and transparent, making it a valuable tool for decision-makers. While healthcare is one application, the framework’s flexibility allows it to be adapted to a wide range of industries. Future work will focus on validating the framework across different sectors and integrating it into operational workflows to optimize decision-making and outcomes. © 2025 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Frequency Domain-based Matching and Integration Network for Multi-contrast MRI Reconstruction

Frequency Domain-based Matching and Integration Network for ...

引用

Image Processing, Computer Vision and Machine Learning (ICICML), International Conference on

作者： Lijuan Wang Xiaozhi Zhang Dongping Xiong Lijun Ouyang School of Computing/Software University of South China Hengyang China School of Electrical Engineering University of South China Hengyang China

Multi-contrast magnetic resonance (MR) images are crucial for diagnosing diseases and analysis in clinical practice, yet their acquisition often entails long scanning procedures. Recently, exploiting shared information among multi-contrast MR images yields favorable results in their reconstruction. However, most studies simply concatenate the multi-contrast information without effective matching and fusion mechanisms to mitigate the impact of redundant features. Furthermore, the feature extraction backbone networks tend to lose high-frequency details like textures during downsampling. To promote the development of this research field, we develop a matching and fusion framework for multi-contrast MR reconstruction and design a feature extractor that explores frequency-domain information. Firstly, our feature extractor captures both long-range and short-range dependencies in the target and reference images, acquiring high- and low-frequency feature information. Then, we introduce a progressive feature matching mechanism to identify features in the reference image that resemble those in the target. Additionally, we propose an innovative fusion strategy that leverages non-local Fourier self-attention and cross-attention mechanisms to precisely combine the semantic feature from multi-contrast MR images. Extensive experiments validate the superior performance of our method in multi-contrast MR image reconstruction, surpassing other state-of-the-art methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Flexible Memory Rotation (FMR): Rotated Representation with Dynamic Regularization to Overcome Catastrophic Forgetting in Continual Knowledge Graph Learning

Flexible Memory Rotation (FMR): Rotated Representation with ...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Zhu, Lijing Jeon, Dong Hyun Sun, Wenbo Yang, Li Xie, Yixin Niu, Shuteng Bowling Green State University Computer Science Department Bowling Green United States University of Michigan Transportation Research Institute Ann Arbor United States University of North Carolina at Charlotte Computer Science Department Charlotte United States Kennesaw State University College of Computing and Software Engineering Marietta United States

ISBN: (纸本)9798350362480

As a special type of Knowledge Graph (KG), Continual Knowledge Graph Learning (CKGL) plays a pivotal role in various areas such as recommendation systems, search engines, and personalized services, where knowledge dynamically evolves. A challenging problem in CKGL is catastrophic forgetting, where models forget previously learned knowledge upon being trained on new data. To overcome the challenge, this study proposes Flexible Memory Rotation (FMR), a dual-level regularization technique that focuses on both parameter level and structural level. Our idea is inspired by the natural human learning process, which tends to memorize correctly learned knowledge, leverage the learned to acquire new knowledge, and refine incorrectly learned knowledge with newly obtained information. Commonly, existing regularization-based methods fail to mimic this human nature by having a fixed constraint strategy for all model parameters. To this end, the proposed FMR offers flexible constraints based on qualities of learned knowledge evaluated by the Fisher Information Matrix (FIM). Additionally, we identified a limitation of FIM in CKGL, which is the assumption of independence of time steps does not always hold. To overcome this, FMR rotates the parameter space to diagonalize the FIM. This work has four major contributions: 1) develop a novel regularization technique, FMR, a flexible regularization technique, 2) reveal the unexpected failure of FIM in CKGL and provides an easy remedy via parameter space rotation, 3) the comparison experiments on four benchmark datasets designed for CKGL demonstrates improvement over various state-of-the-art (SOTA) CKGL models, and 4) a comprehensive ablation study investigates each component of the proposed model. The source code is available at https://***/lijingzhu1/FMR. © 2024 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Chimp Optimization Algorithm and Generalized Normal Distribution Algorithm with Opposition-Based Learning Strategy for Solving Data Clustering Problems

arXiv

引用

arXiv 2023年

作者： Boroujeni, Sayed Pedram Haeri Pashaei, Elnaz School of Computing Clemson University ClemsonSC United States Department of Software Engineering Istanbul Aydin University Istanbul Turkey

This paper is concerned with data clustering to separate clusters based on the connectivity principle for categorizing similar and dissimilar data into different groups. Although classical clustering algorithms such as K-means are efficient techniques, they often trap in local optima and have a slow convergence rate in solving high-dimensional problems. To address these issues, many successful metaheuristic optimization algorithms and intelligence-based methods have been introduced to attain the optimal solution in a reasonable time. They are designed to escape from a local optimum problem by allowing flexible movements or random behaviors. In this study, we attempt to conceptualize a powerful approach using the three main components: Chimp Optimization Algorithm (ChOA), Generalized Normal Distribution Algorithm (GNDA), and Opposition-Based Learning (OBL) method. Firstly, two versions of ChOA with two different independent groups' strategies and seven chaotic maps, entitled ChOA(I) and ChOA(II), are presented to achieve the best possible result for data clustering purposes. Secondly, a novel combination of ChOA and GNDA algorithms with the OBL strategy is devised to solve the major shortcomings of the original algorithms. Lastly, the proposed ChOAGNDA method is a Selective Opposition (SO) algorithm based on ChOA and GNDA, which can be used to tackle large and complex real-world optimization problems, particularly data clustering applications. In this study, eight benchmark datasets including five datasets of the UCI machine learning repository and three challenging shape datasets are used to investigate the general performance of the proposed method. The results are evaluated against seven popular meta-heuristic optimization algorithms and eight recent state-of-the-art clustering techniques. Experimental results illustrate that the proposed work significantly outperforms other existing methods in terms of the achievement in minimizing the Sum of Intra-Cluster Distan

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Magnetic resonance image reconstruction based on two-stage progressive network

Magnetic resonance image reconstruction based on two-stage p...

引用

Electronic Information engineering and Computer Communication (EIECC), International Conference on

作者： Lingyun Shen Xiaozhi Zhang Dongping Xiong Lijun Ouyang School of Electrical Engineering University of South China Hengyang China School of Computing/Software University of South China Hengyang China

In clinical applications, medical image reconstruction is crucial for extracting complementary information and restoring image quality. However, existing deep learning-based reconstruction methods suffer from the following two problems: first, they ignore the frequency-domain information specific to magnetic resonance imaging, and second, most of the traditional networks are single-end-to-end networks, which lack effective attention to feature information. We suggest a unique two-stage progressive network (TSPNet) to solve these issues. Our approach employs two lightweight Unets as subnetworks for progressive recovery, and the network parts are carefully designed. During the feature extraction phase, we creatively create a Dual Domain Feature Extraction Module (DDFEM) to effectively extract rich features. Subsequently, we create a hybrid attention module (HAM) to effectively process and integrate the retrieved data. Experimental results show that our model outperforms other state-of-the-art methods and reconstructs images with stronger texture information and superior visual quality.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Earliest-Arrival Route: A Global Optimized Communication for Networked Control Systems 8th

Earliest-Arrival Route: A Global Optimized Communication for...

引用

8th EAI International Conference on Advancement of Science and Technology, ICAST 2020

作者： Rajan, M. Sundar Arunkumar, J.R. Anusuya, R. Mesfin, Abraham Faculty of Electrical and Computer Engineering Arbaminch University Arba Minch Ethiopia Faculty of Computing and Software Engineering Arbaminch University Arba Minch Ethiopia

ISBN: (纸本)9783030806200

As per the present switched system theorem, data package-drop ratio decides the strength of NCSs. Through the focal point principle to offer small data package-drop ratio for dispersed NCS under parameter transmission environment, the communication improvement issue is described from universal transmission aspect. Initially, system stimulated break is illustrated function associated with time can be calculated using auto-regressive-moving-average (ARMA) pattern, and the upcoming break is forecasted by the obtained break samples. Subsequently, the widespread referred shortest route of time dependent network(SPTDN) algorithm is described and enhanced by means of introducing waiting-time previous to data packages are dispatched to subsequent unit in a route, in order to get quickly coming time than SPTDN. Subsequently, the concept of earliest-arrival time (EAT) is suggested to calculate the excellence of transmission route, the connected EAT algorithm is intended to look for upcoming EAT and the equivalent route since universal system vigorously. Advantage from the active EAT route, data packages reach at target as early as feasible, it reduces data package-drop ratio. Therefore, EAT route make sure the steadiness of shared NCSs from the opinion of worldwide system. At last, the experiment demonstrates the strength and efficiency of the suggested method. © 2021, ICST Institute for Computer Sciences, Social Informatics and Telecommunications engineering.

关键词： Networked control systems

来源：评论

学校读者我要写书评

暂无评论

FedRec++: Lossless Federated Recommendation with Explicit Feedback 35

FedRec++: Lossless Federated Recommendation with Explicit Fe...

引用

35th AAAI Conference on Artificial Intelligence, AAAI 2021

作者： Liang, Feng Pan, Weike Ming, Zhong National Engineering Laboratory for Big Data System Computing Technology College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China

ISBN: (纸本)9781713835974

With the marriage of federated machine learning and recommender systems for privacy-aware preference modeling and personalization, there comes a new research branch called federated recommender systems aiming to build a recommendation model in a distributed way, i.e., each user is represented as a distributed client where his/her original rating data are not shared with the server or the other clients. Notice that, besides the sensitive information of a specific rating score assigned to a certain item by a user, the information of a user’s rated set of items shall also be well protected. Some very recent works propose to randomly sample some unrated items for each user and then assign some virtual ratings, so that the server can not identify the scores and the set of rated items easily during the server-client interactions. However, the virtual ratings assigned to the randomly sampled items will inevitably introduce some noise to the model training process, which will then cause loss in recommendation performance. In this paper, we propose a novel lossless federated recommendation method (FedRec++) by allocating some denoising clients (i.e., users) to eliminate the noise in a privacy-aware manner. We further analyse our FedRec++ in terms of security and losslessness, and discuss its generality in the context of existing works. Extensive empirical studies clearly show the effectiveness of our FedRec++ in providing accurate and privacy-aware recommendation without much additional communication cost. Copyright © 2021, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Recommender systems

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：