检索结果-内蒙古大学图书馆

Mexican international conference on Computer Science (ENC)

作者： Tablada-Dominguez, Antonio Munoz, Mirna Ocharan-Hernandez, Jorge Octavio Sanchez-Garcia, Angel J. Ctr Inmvestigac Matemat AC Zacatecas Zacatecas Mexico Univ Veracruzana Xalapa Veracruz Mexico

ISBN: (纸本)9798350393156

The AI era established significant challenges for software developers, especially those working on machine learning (ML)-based software. This article presents the findings of a systematic literature review (SLR) focused on identifying software engineering practices for ML-based software development. We identified 16 primary studies highlighting the challenges scientists face in lacking software engineering training when developing ML-based software. The results emphasize the importance of documentation, standardized processes, and skills acquisition to overcome these challenges effectively in the AI era.

关键词： software engineering software development software processes software best practices machine learning artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Preface to the special issue on engineering of computer-based systems

引用

international JOURNAL ON software TOOLS FOR TECHNOLOGY TRANSFER 2025年第1期27卷 1-3页

作者： Kofron, Jan Margaria, Tiziana Seceleanu, Cristina Charles Univeristy Prague Czech Republic Univ Limerick Limerick Ireland LERO Limerick Ireland Malardalen Univ Vasteras Sweden

This special issue contains nine extended and rigorously peer-reviewed papers selected from those originally presented at ECBS 2023, the 8th international conference on engineering of Computer-Based Systems, held at M & auml;lardalen University, Sweden, October 16-18, 2023, under the theme "engineering for Responsible AI". The included papers represent innovative contributions addressing critical aspects of responsible artificial intelligence and integrated engineering practices. These contributions span from formal verification and security analyses of IoT protocols and federated learning frameworks to machine learning-based simulations and predictions in hardware and software systems. The selection also includes work on automata learning techniques for protocol compliance, continuous integration approaches for neural network-based autonomous systems, assertion usage in software testing, language-driven engineering for code generation, and the integration of IoT backends in digital twin infrastructures. Together, these papers showcase recent advances, offering valuable insights into the rigorous integration of modern technologies within complex, computer-based systems.

关键词： Security of IoT machine learning Automata learning Testing Digital twins

来源：评论

学校读者我要写书评

暂无评论

software Fault Prediction Using ML Algorithms 5th

Software Fault Prediction Using ML Algorithms

引用

5th international conference on Data Science, machine learning and Applications

作者： Neha, T. Anusha, M. Anusha, D. G Narayanamma Inst Technol & Sci Hyderabad Telangana India

ISBN: (纸本)9789819780334;9789819780310;9789819780303

software fault prediction (SFP) is becoming increasingly important in software engineering, especially in service-oriented systems (SOS). This study investigates the effectiveness of using source code for fault prediction in SOS. It uses supervised machine learning algorithms such as random forest, decision tree, and support vector machine to improve error prediction accuracy. Feature extraction is used for more accurate analysis. The study highlights the strengths and weaknesses of these algorithms, providing insights into the prediction of malicious software in SOS. It aims to provide high-performance and reliable software architecture, and advance fault prediction models in SFP.

关键词： Random forest software bug prediction Support vector machines Long short term memory Neural networks

来源：评论

学校读者我要写书评

暂无评论

SCAM-GAN: Generating brain MR images from CT scan data based on CycleGAN combined with attention module

SCAM-GAN: Generating brain MR images from CT scan data based...

引用

2023 international conference on software engineering and machine learning, conf-seml 2023

作者： Liu, Songming School of Information Management and Engineering Artificial Intelligence Zhejiang University of Finance & Economics Zhejiang310018 China

The diagnosis of epilepsy often depends heavily on Magnetic Resonance Imaging (MRI). Unfortunately, the utilization of MRI is constrained, due to its expensive price and lengthy operating times. More significantly, certain people with claustrophobia or cardiac pacemakers are not candidates for MRI owing to the risk of harm. Computed tomography (CT) images, in comparison, are considerably faster, cheaper, and free from the same restrictions. As opposed to conventional medical imaging synthetic techniques, which rely on abundant unpaired data or a minimum number of paired data, the proposed method in this study utilizes unpaired training data to predict an MR picture from a CT scan. This method overcomes tight registration problem in paired training and reduces the challenge of context mismatch in unpaired training. It is possible to convert 2D brain MR pictures from CT scans using the Spatial & Channel Attention Mechanism-Generative Adversarial Network (SCAM-GAN) that includes cycle-consistent and adversarial losses. The superiority of the proposed strategy was established by quantitative comparisons versus separate training approaches which are paired and unpaired. © 2023 Institute of Physics Publishing. All rights reserved.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Traffic sign detection based on classic visual recognition models

Traffic sign detection based on classic visual recognition m...

引用

2023 international conference on software engineering and machine learning, conf-seml 2023

作者： Yang, Jirui College of Engineering Computing and Cybernetics The Australian National University Canberra Australia

This paper aims to assess the effectiveness of various object detection high-level architectures, including Faster R-CNN, R-FCN, SSD, and YOLO, in recognizing traffic signs. Since traffic sign recognition is a critical element of driver-assistance and autonomous driving systems, this study explores how five distinct architectures for feature extracting (ResNet V1 50, ResNet V1 101, Inception V2, Inception-ResNet-V2, and Darknet-19) can enhance the performance of these meta-architectures. To evaluate the models, the authors fine-tuned a pre-trained object recognition algorithm on the German Traffic Sign Detection Benchmark (GTSDB) dataset. While the Faster R-CNN Inception Resnet v2 proved to be the most accurate among the models, the unbalanced distribution of mandatory, prohibitory, and danger signs in the GTSDB dataset may have caused biased results in object detection. Overall, the result obtained from this study can help improve the accuracy of driver-assistance and autonomous driving systems by providing insights into the performance of various object detection models in recognizing traffic signs. © 2023 Institute of Physics Publishing. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Combining feature pyramid and CycleGAN for image generation

Combining feature pyramid and CycleGAN for image generation

引用

2023 international conference on software engineering and machine learning, conf-seml 2023

作者： Li, Yuqi Lu, Jiren Meng, Xiangyu School of Artificial Intelligence Tiangong University Tianjin300387 China School of Statistics and Information Shanghai University of International Business and Economics Shanghai201620 China College of Software Tianjin Normal University Tianjin300384 China

Image generation has always been a hot topic in computer vision community, which aims to learn the data distribution from a give image dataset and then generates new images obeying this distribution. Thanks to the rapid development of convolutional neural networks, breakthroughs have been made in the accuracy and speed of image generation. Currently, image generation is mainly based on the framework for generating countermeasures networks. However, limited by the quality of features, the generated images still have problems such as edge blurring, which restricts large-scale practical applications. In this paper, we propose an image generation algorithm that combines feature pyramid network (FPN) and CycleGAN. Specifically, the FPN containing two upsampling and feature fusion operations is added after the residual blocks and before the decoder of the CycleGAN generator, which can help the generator produce more detailed images and learn from features at different scales. Qualitative and quantitative experimental results show that the CycleGAN model with FPN outperforms the original CycleGAN model in terms of image generation. © 2023 Institute of Physics Publishing. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Edge-centric energy efficient LSTM with federated learning model based computation offloading for IoT applications

Edge-centric energy efficient LSTM with federated learning m...

引用

2023 international conference on software engineering and machine learning, conf-seml 2023

作者： Selvamani, P. Anitha, R. Department of Computer Science and Engineering Sri Venkateswara Colleg of Engineering Tamil Nadu Sriperumbudur India

A more recent innovation to support cloud computing is edge computing that can address the deficiency of the existing centralised cloud computing paradigm and bring compute and storage resources closer to devices. Edge computing isn't always the same as conventional cloud computing. Edge computing isn't always the same as conventional cloud computing. It is a new computational model that performs computation at the network edge. Its crucial idea is to bring computing in the way of the information’s source. Task scheduling is the process allocating incoming requests (tasks) using a specific method to make the best use of the resources for the needed process. Users of services must submit their requests online because cloud computing is the technology used to deliver services through the internet. This paper aims to provide an Edge centric LSTM model with Federated learning for Data storage and model training that will take place on powerful edge servers in distributed ML methods. a massively distributed tasks requiring Processing on-site, coordinating the task remotely and execution are carried out in collaboration with number of multiple edge nodes and the distant cloud infrastructure. As a result, improved computation offloading and networking tradeoffs are possible to achieve low latency, large bandwidth take place on powerful edge servers. © 2023 Institute of Physics Publishing. All rights reserved.

关键词： Computation offloading

来源：评论

学校读者我要写书评

暂无评论

YOLO sparse training and model pruning for street view house numbers recognition

YOLO sparse training and model pruning for street view house...

引用

2023 international conference on software engineering and machine learning, conf-seml 2023

作者： Zhang, Ruohao Lu, Yijie Song, Zhengfei Software Institute Nanjing University China School of Artificial Intelligence Nanjing University China College of Electronic and Information Engineering Tongji University China

This paper proposes a YOLO (You Only Look Once) sparse training and model pruning technique for recognizing house numbers in street view images. YOLO is a popular object detection algorithm that has achieved state-of-the-art performance in various computer vision tasks. However, its large model size and computational complexity limit its deployment on resource-constrained devices such as smartphones and embedded systems. To address this issue, we use a sparse training technique that trains YOLO with L1 norm regularization to encourage the network to learn sparse representations. This results in a significant reduction in the number of parameters and computation without sacrificing accuracy. Furthermore, we apply a model pruning technique to the sparse-trained model to reduce the model size and computation. We evaluate our proposed method on the SVHN (Street View House Numbers) dataset. We show that it performs comparably to the original YOLO model while reducing the model size by 5% and the computation time by 7%. Overall, our proposed YOLO sparse training and model pruning technique provides an effective solution for deploying YOLO-based object detection models on resource-constrained devices. © 2023 Institute of Physics Publishing. All rights reserved.

关键词： Houses

来源：评论

学校读者我要写书评

暂无评论

Impact of machine learning on software Development Life Cycle 18

Impact of Machine Learning on Software Development Life Cycl...

引用

18th international conference on Evaluation of Novel Approaches to software engineering (ENASE)

作者： Navaei, Maryam Tabrizi, Nasseh East Carolina Univ Dept Comp Sci East 5th St Greenville NC 27858 USA

ISBN: (纸本)9789897586477

This research concludes an overall summary of the publications so far on the applied machine learning (ML) techniques in different phases of software Development Life Cycle (SDLC) that includes Requirement Analysis, Design, Implementation, Testing, and Maintenance. We have performed a systematic review of the research studies published from 2015-2023 and revealed that software Requirements Analysis phase has the least number of papers published;in contrast, software Testing is the phase with the greatest number of papers published.

关键词： software engineering software Development Life Cycle Artificial Intelligence machine learning machine learning Algorithms

来源：评论

学校读者我要写书评

暂无评论

Ghost Echoes Revealed: Benchmarking Maintainability Metrics and machine learning Predictions Against Human Assessments 40

Ghost Echoes Revealed: Benchmarking Maintainability Metrics ...

引用

40th international conference on software Maintenance and Evolution

作者： Borg, Markus Ezzouhri, Marwa Tornhill, Adam CodeScene Malmo Sweden Lund Univ Lund Sweden Univ Clermont Auvergne Clermont Ferrand France

ISBN: (纸本)9798350395693;9798350395686

As generative AI is expected to increase global code volumes, the importance of maintainability from a human perspective will become even greater. Various methods have been developed to identify the most important maintainability issues, including aggregated metrics and advanced machine learning (ML) models. This study benchmarks several maintainability prediction approaches, including State-of-the-Art (SotA) ML, SonarQube's Maintainability Rating, CodeScene's Code Health, and Microsoft's Maintainability Index. Our results indicate that CodeScene matches the accuracy of SotA ML and outperforms the average human expert. Importantly, unlike SotA ML, CodeScene also provides end users with actionable code smell details to remedy identified issues. Finally, caution is advised with SonarQube due to its tendency to generate many false positives. Unfortunately, our findings call into question the validity of previous studies that solely relied on SonarQube output for establishing ground truth labels. To improve reliability in future maintainability and technical debt studies, we recommend employing more accurate metrics. Moreover, reevaluating previous findings with Code Health would mitigate this revealed validity threat.

关键词： software engineering maintainability software metrics benchmarking machine learning

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：