检索结果-内蒙古大学图书馆

FedPPO: Reinforcement Learning-Based Client Selection for Federated Learning With Heterogeneous Data

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Zhao, Zheyu Li, Anran Li, Ruidong Yang, Lei Xu, Xiaohua University of Science and Technology of China School of Computer Science and Technology Hefei Anhui230026 China Kanazawa University Institute of Natural Sciences Ishikawa9201192 Japan South China University of Technology School of Software Engineering Guangzhou510006 China

Federated Learning (FL) enables multiple data owners to jointly train a machine learning model, which can improve joint environmental cognitive capability without disclosing their private local data. In FL systems, local clients are often heterogeneous, characterized by different data distribution forms and varying levels of noise patterns. This poses challenges to the convergence and test accuracy of the learned global model. Existing client selection methods failed to effectively address the data heterogeneity problem while considering the impacts of correlations between clients. In this work, we propose an efficient and adaptive client selection scheme with the Federated Proximal Policy Optimization (FedPPO). It utilizes a noisy client filtering method that leverages the principal component analysis and hierarchical clustering to reduce the action search space and accelerate the convergence of Deep Reinforcement Learning (DRL). It then evaluates each client's contribution by jointly using the accuracy of global model and local models and take the contribution score as the reward. Finally, it leverages a DRL algorithm, PPO, to adaptively select clients with high contributions to participate in training. Theoretical analysis shows that it is guaranteed to converge in an efficient manner. Experimental results on four real-world datasets show that FedPPO significantly outperforms existing approaches of constructing global models with faster convergence speed and higher test accuracy. © 2015 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Chinese Speech Emotion Recognition Based on Pre-Trained Model

Chinese Speech Emotion Recognition Based on Pre-Trained Mode...

引用

Big Data & Artificial Intelligence & software engineering (ICBASE), International Conference on

作者： Sukui Xu Nan Jiang Shun Lian Jingmin Pan College of Computer and Software Engineering Anhui Institute of Information Technology Wuhu China School of General Education Fujian Polytechnic of Information Technology Fuzhou China School of Computer Science and Technology Anhui University Hefei China

ISBN: (数字)9798331506612

ISBN: (纸本)9798331506629

This paper proposes a method based on large-scale speech pre-trained models for the task of Chinese speech emotion recognition. Similar to the transfer learning approach in image classification task, speech pre-trained model is used as a fixed feature extractor to extract features from speech data, followed by a fully connected network for classification. Experimental results show that, on the seven-class speech emotion recognition task of the Chinese M3ED dataset, the proposed method achieves classification accuracies of 0.506 and 0.560 using base and large sized pre-trained models respectively, surpassing the baseline accuracy of 0.454 for the ECAPA-TDNN model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Progression of Competencies and Dispositions: How do Maturing software Engineers Compare?

The Progression of Competencies and Dispositions: How do Mat...

引用

Frontiers in Education (FIE) Conference

作者： Tony Clear Alison Clear Department of Computer Science and Software Engineering Auckland University of Technology Auckand New Zealand School of Computing Eastern Institute of technology Auckland New Zealand

ISBN: (数字)9798350351507

ISBN: (纸本)9798350363067

This research paper describes and extends the outcomes from an in- depth study investigating the difference in the expected skills requirements from junior software engineers to senior software engineers, and reflections on the findings from that study. It is a given that senior software engineers have more experience and skills than junior software engineers. However, a focus on their differing competencies and dispositions provides an enhanced mechanism for comparison. Gaps were identified in assessing “professional knowledge” as categorized by the IEEE/ACM Computing Curriculum Overview Report (CC2020), and in assessing “dispositions”. It appeared that the specific scenario of comparing the expected competencies between junior and senior software engineers, tested the framework for assessing competencies developed in the CC2020 project and applied in its mapping to the IEEE/ACM computer Science (CS2013) approved curriculum. In this study into the difference between Junior and Senior software Engineers, an initial review of relevant literature was conducted. The review found that research analyzing job requirements for software engineers of different levels was limited; “experience” as a keyword was seldom mentioned; and a common distinction was made between “soft” and “hard” skills - the latter being skills that were “technical”, such as programming languages, frameworks, libraries, and tools, whereas soft skills referred to skills such as personality traits, attitudes, and teamwork skills. In our extension of that work the notion of soft skills was unpacked into professional skills and dispositions. The process of mapping from the CC2020 competency framework to the CS2013 curriculum had deliberately modelled how to represent a competency-based rather than a knowledge-based curriculum. The critical deficiency identified here was the limitation imposed by adopting a skills framework based on the cognitive taxonomy, and thereby unwittingly omitting the crucial compa

关键词： computer science Vocabulary computer languages Reviews Taxonomy Knowledge based systems software Reflection Libraries Teamwork

来源：评论

学校读者我要写书评

暂无评论

Offline Teaching Method in Blended Teaching mode of Improving the Innovation Ability of Students in IT Majors 22

Offline Teaching Method in Blended Teaching mode of Improvin...

引用

7th International Conference on Information and Education Innovations, ICIEI 2022

作者： Huang, Min Yuan, Hua Sun, Bo School of Software engineering South China University of Technology Guangzhou China School of Computer science and engineering South China University of Technology Guangzhou China Research Institute of international services outsourcing GuangDong university of Foreign studies Guangzhou China

ISBN: (纸本)9781450396196

At present, all kinds of innovative applications of artificial intelligence, big data and other technologies in various industries and fields make the demands for the cultivation of innovative ability of IT talents more urgent. Aiming at the problem of insufficient cultivation of students' innovative ability in the existing blended teaching mode, this paper puts forward a theoretical teaching method based on four-stage and problem-oriented in the offline class, and a practical teaching method based on "entrepreneurial thinking". In the former teaching method, by dividing the explanation of a concrete theoretical knowledge into four stages: "concept", "problem", "application"and "innovation", and at the same time, designing some progressive problems in each stage to guide students to deepen their knowledge understanding and inspire their innovative thinking around the application and innovation of knowledge, and an example has been given to well illustrate the method. Meanwhile, in the latter teaching method, this paper also presents a practical teaching method based on "entrepreneurial thinking"to optimize the practical teaching links of the courses in IT major, in which the project-based teaching method and entrepreneurial thinking are adopted , and the practical process is divided into several stages as "demand analysis and topic selection", "scheme design", "scheme implementation"and "application and promotion", At each stage, some concrete guiding measures and examination and evaluation methods have been designed to effectively stimulate students' innovative thinking and innovative ability specific. Through the application of the above methods to the teaching practices in the "computer network"course, it is proved that this method plays a promoting role in the development of students' innovation and practical abilities. © 2022 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Algorithms for Intrusion Detection Performance Evaluation and Comparative Analysis 4

Machine Learning Algorithms for Intrusion Detection Performa...

引用

4th International Conference on Smart Electronics and Communication, ICOSEC 2023

作者： Irfan, B. Md. Poornima, V. Mohana Kumar, S. Aswal, Upendra Singh Krishnamoorthy, N. Maranan, Ramya Nalsar University of Law Department of Information Technology Telengana Hyderabad500101 India VISTAS Pallavaram Department of Computer Science Chennai India Ramaiah Institute of Technology Department of Computer Science and Engineering Karnataka Bengaluru560054 India Graphic Era Deemed to Be University Department of Computer Science and Engineering Uttarakhand Dehradun248002 India Vellore Institute of Technology Vellore Campus Department of Software Systems and Engineering Tamilnadu Vellore632014 India Saveetha School of Engineering Simats Department of Research and Innovation Tamil Nadu Chennai600124 India

ISBN: (纸本)9798350300888

The security of computer networks is increasingly difficult to maintain due to the rising complexity and frequency of cyber-attacks. Important tools for finding and neutralizing these dangers are intrusion detection systems. This study sets out to do a thorough examination and comparison of the efficacy of several machine learning algorithms for use in intrusion detection. This research study evaluates the efficacy of several machine learning algorithms in correctly categorizing instances of network traffic as normal or invasive via extensive experiments performed on representative datasets. Algorithms like random forests, decision trees, SVMs, DL models and NNs are all being tested and rated. Effectiveness is measured and compared using a variety of performance indicators including accuracy, recall, precision, false positive rate, and F1-score. The results of this study emphasize the potential of deep learning models and Random Forests for use in intrusion detection and add to the body of knowledge around machine learning methods for this task. Professionals in the field of network security might use the results to their advantage when building intrusion detection systems. Future research areas are also mentioned, which will hopefully lead to even greater improvements in the field and safer, more reliable intrusion detection systems. © 2023 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Integrating Design Thinking into software Process Improvement 4

Integrating Design Thinking into Software Process Improvemen...

引用

4th International Conference on Emerging Trends in Networks and computer Communications, ETNCC 2024

作者： Khalid, Anum Hashmi, Amber Sarwar Kiani, Azaz Ahmed Muzammal, Syeda Mariam Jhanjhi, N.Z. Bibi, Ruqia Kumar, Sandeep Foundation University Department of Software Engineering Islamabad Pakistan Rawalpindi Women University Department of Software Engineering Rawalpindi Pakistan National University of Modern Languages Rawalpindi Pakistan School of Computer Science Taylor's University Faculty of Innovation and Technology Subang Jaya Malaysia University Institute of Information Technology Pir Mehr Ali Shah Arid Agriculture University Rawalpindi Pakistan Lovely Professional University Division of Research and Development Phagwara144411 India KKR and KSR institute of technology and sciences Andhra Pradesh Guntur India

ISBN: (纸本)9798350353266

The need for software improvement is growing widely due to the demand for technology development. The issues that arise by expanding the software processes in the workplace are expense, lack of understanding, time constrained, and lack user experience. This research paper focuses on proposing a framework for integrating design thinking into software process improvement. The secret to project success may lie in fusing methodological and reusable approaches with human-centric techniques. A mixed-methods research strategy is proposed to compile thorough information and reach conclusions. In addition, it incorporates components of both qualitative and quantitative research (secondary research and descriptive analysis). The assumption is that the existing model significantly affects the user satisfaction variable. This indicates a statistically significant association between the individual's experience (number of years) and design thinking integration into the software process improvement. The variables in the proposed model and the dependent variable user satisfaction accounts for 6.1 © 2024 IEEE.

关键词： software design

来源：评论

学校读者我要写书评

暂无评论

Data Centric Blockchain Based Evaluation Approach to Analyze E-Commerce Reviews Using Machine and Deep Learning Techniques 20

Data Centric Blockchain Based Evaluation Approach to Analyze...

引用

20th International computer Conference on Wavelet Active Media technology and Information Processing, ICCWAMTIP 2023

作者： Acheampong, Edward Mensah Zhou, Shijie Liao, Yongjian Atandoh, Peter Addo, Daniel Antwi-Boasiako, Emmanuel Nkrumah, Rose Aggrey, Esther Stacy E.B. Appiah-Twum, Michael School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China Ghana Institute of Management and Public Administration Department of Computer Science Ghana University of Cape Coast Department of Basic Education Ghana School of Resource and Environment University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350318982

Machine and deep learning techniques are now essential for many different kinds of solutions in many areas, applications, and businesses. Analyzing e-commerce reviews using machine and deep learning techniques can motivate a company to make more wise decisions regarding the quality and services it offers. But issues can occur if there is insufficient high-quality data for the training, scalability, and upkeep of learning models. To address these important issues, we suggest a data-centric learning architecture that makes use of a blockchain and Interplanetary File System (IPFS)-based storage. Our architecture provides secure and cost-effective data storage and an incentive mechanism. Our experiment shows that as the dataset gets bigger, the Naïve Bayes (NB) and the Support Vector Machine (SVM) models get 2% more accurate, the Convolutional Neural Network (CNN) model gets 4% more accurate, and the Long Short-Term Memory (LSTM) model gets 3 % more accurate. The deep learning models exhibited higher accuracy rates than the machine learning models because hidden layers of deep learning possess the ability to extract intricate syntactic features from the review text. © 2023 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

Feature selection and prediction of small-for-gestational-age infants

引用

Journal of Ambient Intelligence and Humanized Computing 2024年第3期15卷 1881-1895页

作者： Li, Jianqiang Liu, Lu Zhou, MengChu Yang, Ji-Jiang Chen, Shi Liu, HuiTing Wang, Qing Pan, Hui Sun, ZhiHua Tan, Feng School of Software Engineering Beijing University of Technology Beijing China Department of Electronical and Computer Engineering New Jersey Institute of Technology NewarkNJ United States Tsinghua University Beijing China Department of Endocrinology Peking Union Medical College Hospital Beijing China Department of Internal Medicine Peking Union Medical College Hospital Beijing China Beijing Chaoyang District Maternal and Child Healthcare Hospital Beijing China Chinese Center For Disease Control And Prevention Beijing China

The small-for-gestational-age (SGA) condition often causes serious problems. Therefore, identifying the risk factors for SGA is important. Traditional statistical methods such as stepwise logistic regression (LR) have been widely utilized to discover possible risk factors. However, other feature selection methods from machine learning field have rarely been employed for the task. In this paper, a comparison of five feature selection methods from both fields for SGA risk factors analysis is conducted for the first time. To evaluate their performance, four classification algorithms are used to construct SGA prediction models. The evaluation criteria are precision and the area under the receiver operator characteristic curve. Stepwise LR achieves the best performance among the five feature selection methods, because it conducts both a univariate significance test and a model significance test, which make it more suitable for handling the complex relations among features. The top 20 features selected by each feature selection method and the 27 features selected by four or five of them could assist physicians to revise traditional SGA evaluation models. Ensemble method is also exploited to build effective SGA prediction models based on the feature subsets, which is indeed superior compared with the individual ones shown in the results. © Springer-Verlag GmbH Germany, part of Springer Nature 2018.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Location Intelligence Based Smart E-Commerce Platform for Residential Real-Estate Industry 3

Location Intelligence Based Smart E-Commerce Platform for Re...

引用

3rd International Conference on Smart Electronics and Communication, ICOSEC 2022

作者： Balasooriya, Tharindu Kavishka, Lakindu Dananjana, Isini Yumna, Maryam Thelijjagoda, Samantha Attanayaka, Buddhima Marasinghe, Lanil Sri Lanka Institute of Information Technology Department of Computer Science and Software Engineering Malabe Sri Lanka Sri Lanka Institute of Information Technology Sliit Business School Malabe Sri Lanka Department of Computer Systems Engineering Sri Lanka Institute of Information Technology Malabe Sri Lanka Sysco Labs Colombo Sri Lanka

ISBN: (数字)9781665497640

ISBN: (纸本)9781665497640

Making the decision to purchase or invest in real estate can be a very crucial process due to its high financial risk. The purchasing decision of residential real estate properties can be even more decisive because, apart from the financial risk, the choice of a property can have a great impact on the future lifestyle of the buyer. When considering residential real estate, one major factor to be considered is the property location. This research sought to determine the applicability of modern technologies such as location intelligence and machine learning in the development of an e-commerce system that may assist users in making optimal residential real estate location decisions. Third-party web APIs were used to obtain location data, and as a result of the study, methodologies were defined to convert location data into meaningful insights by using statistical methods like weighted sum and the analytical hierarchy process. The functionalities in the proposed system have been designed, considering the roles of both buyers and sellers in the real estate business. In the proposed system, the location quality index framework provides overall insights on the location, the personalized insights and alternative locations are generated via the personal preference-based suitability analyzer and the price prediction system provides the current and future price fluctuations. The usefulness of image processing technologies and machine learning for making the sellers' journey easier on a real estate platform has also been assessed. © 2022 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Keeping Pace with Ever-Increasing Data: Towards Continual Learning of Code Intelligence Models

arXiv

引用

arXiv 2023年

作者： Gao, Shuzheng Zhang, Hongyu Gao, Cuiyun Wang, Chaozheng School of Computer Science and Technology Harbin Institute of Technology Shenzhen China School of Big Data and Software Engineering Chongqing University China

—Previous research on code intelligence usually trains a deep learning model on a fixed dataset in an offline manner. However, in real-world scenarios, new code repositories emerge incessantly, and the carried new knowledge is beneficial for providing up-to-date code intelligence services to developers. In this paper, we aim at the following problem: How to enable code intelligence models to continually learn from ever-increasing data? One major challenge here is catastrophic forgetting, meaning that the model can easily forget knowledge learned from previous datasets when learning from the new dataset. To tackle this challenge, we propose REPEAT, a novel method for continual learning of code intelligence models. Specifically, REPEAT addresses the catastrophic forgetting problem with representative exemplars replay and adaptive parameter regularization. The representative exemplars replay component selects informative and diverse exemplars in each dataset and uses them to retrain model periodically. The adaptive parameter regularization component recognizes important parameters in the model and adaptively penalizes their changes to preserve the knowledge learned before. We evaluate the proposed approach on three code intelligence tasks including code summarization, software vulnerability detection, and code clone detection. Extensive experiments demonstrate that REPEAT consistently outperforms baseline methods on all tasks. For example, REPEAT improves the conventional fine-tuning method by 1.22, 5.61, and 1.72 on code summarization, vulnerability detection and clone detection, respectively. Copyright © 2023, The Authors. All rights reserved.

关键词： Cloning

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：