Extracting multiple kinds of knowledge from large ship technical specification texts plays an important role in improving the efficiency of designers in acquiring and applying knowledge. In view of the huge number of ...
详细信息
Machine learning (ML) models have become essential components in software systems across several domains, such as autonomous driving, healthcare, and finance. the robustness of these ML models is crucial for maintaini...
详细信息
ISBN:
(数字)9798400712487
ISBN:
(纸本)9798400712487
Machine learning (ML) models have become essential components in software systems across several domains, such as autonomous driving, healthcare, and finance. the robustness of these ML models is crucial for maintaining the software systems performance and reliability. A significant challenge arises when these systems encounter out-of-distribution (OOD) data, examples that differ from the training data distribution. OOD data can cause a degradation of the software systems performance. therefore, an effective OOD detection mechanism is essential for maintaining software system performance and robustness. Such a mechanism should identify and reject OOD inputs and alert software engineers. Current OOD detection methods rely on hyperparameters tuned with in-distribution and OOD data. However, defining the OOD data that the system will encounter in production is often infeasible. Further, the performance of these methods degrades with OOD data that has similar characteristics to the in-distribution data. In this paper, we propose a novel OOD detection method using the Gini coefficient. Our method does not require prior knowledge of OOD data or hyper-parameter tuning. On common benchmark datasets, we show that our method outperforms the existing maximum softmax probability (MSP) baseline. For a model trained on the MNIST dataset, we improve the OOD detection rate by 4% on the CIFAR10 dataset and by more than 50% for the EMNIST dataset.
the ability to simulate is influenced considerably by the software technology available to the modeler. the access, availability, and use of simulation software are, in turn, affected by cost considerations, geographi...
详细信息
Code quality and maintainability are among under-emphasized and often neglected topics in the curriculum of softwareengineering (SE) in higher *** neglect tends to overlook research findings that demonstrate SE stude...
详细信息
ISBN:
(纸本)9798400704987
Code quality and maintainability are among under-emphasized and often neglected topics in the curriculum of softwareengineering (SE) in higher *** neglect tends to overlook research findings that demonstrate SE students' programming submissions most often exhibit severe code quality issues, which are frequently left unaddressed by the ***, it can result in the softwareengineering curriculum becoming indifferent to the essential requirements of the software development industry, where code quality and maintainability play a crucial role in the software's cost throughout its life cycle. therefore, SE students in higher education should be trained to master the knowledge and skills of writing high-quality code. One possible approach to improving students' understanding of code quality issues is to provide automatically generated formative feedback about the code quality aspects of their programming submissions throughout the code development process. However, while there are tools available for generating automated feedback on the code quality aspects of programming submissions, they often lack a set of theory-driven design principles to underpin the content and presentation of their provided feedback. this lack of theoretical foundation makes it difficult to follow a systematic approach to designing and developing such tools, reasoning about their quality, and evaluating the effectiveness of their generated feedback. To address this lack, this study provides nine contextualized design principles for generating automated formative feedback on code quality. these design principles are rooted in solid educational constructs about feedback and learning dashboards, and empirically validated and contextualized by two focus group sessions consisting of 8 senior SE students and 2 teachers. this approach has resulted in a set of contextualized design principles. these design principles can be used to guide the implementation of tools that provide
In project-based softwareengineering courses, development teams conformed by students explore the context and problem to address in their projects, before proposing any solution. the quality of this exploration, and ...
详细信息
ISBN:
(纸本)9783031601248;9783031601255
In project-based softwareengineering courses, development teams conformed by students explore the context and problem to address in their projects, before proposing any solution. the quality of this exploration, and the resulting output, usually make a difference on how quickly the development team identifies the goal and scope of the product to be developed. this exploration activity is usually complex and time consuming, since it requires dealing with uncertainties and misunderstandings between the development team and real or fictitious stakeholders. For that reason, the early exploration of the context and problem has been identified as a major and recurrent source of problems in software projects conducted in the industry and the academia. this paper presents an interactive visual tool that helps students explore the context and problem to address in project-based softwareengineering courses. the tool was used and evaluated by students from four different courses in two universities. According to the participants, the perceived usability and usefulness of the tool is high, surpassing the students' previous experiences when they used requirements engineering techniques for the same purpose.
In recent years, researchers and software developers have applied software process automation techniques during the execution of projects achieving acceptable results. thus, a considerable variety of process automatio...
详细信息
ISBN:
(纸本)9783031752353;9783031752360
In recent years, researchers and software developers have applied software process automation techniques during the execution of projects achieving acceptable results. thus, a considerable variety of process automation techniques are applied in each of its phases. the objective of this study is to perform a systematic review of the literature on software process automation techniques. the results indicate that automation techniques are consolidated in phases such as testing, continuous integration and model-based softwareengineering. these techniques are being implemented to optimize build time, provide greater product autonomy and rely less and less on software engineers. Some of these solutions are focused on improving the phases of the process, in order to build quality software at lower cost and in less time.
the proceedings contain 256 papers. the special focus in this conference is on Biomedical engineering Systems and Technologies. the topics include: Cross-Domain Transfer Learning for Domain Adaptation in Autism Spectr...
the proceedings contain 256 papers. the special focus in this conference is on Biomedical engineering Systems and Technologies. the topics include: Cross-Domain Transfer Learning for Domain Adaptation in Autism Spectrum Disorder Diagnosis;an Interpretable Machine Learning Model for Meningioma Grade Prediction;approximation of Inertial Measurement Unit Data to Time Series Kinematic Data through Correlation Analysis and Machine Learning;Privacy-Preserving Mortality Prediction in ICUs Using Federated Learning;trends in Drug Prescriptions in the Outpatient Physician Sector in a German Federal State from 2014 to 2023 Using Morbidity Related Groups, Correlations and Partial Correlations;quality Clustering for Reducing the Search Space for Mobile Stroke Unit Allocation;leveraging Cross-Verification to Enhance Zero-Shot Prompting for Care Document Data Extraction;Integrating Gait and Clinical Data with Explainable Artificial Intelligence for Parkinson’s Prediction: the EDAM System;Unveiling Breast Cancer Causes through knowledge Graph Analysis and BioBERT-Based Factuality Prediction;optimizing Blood Transfusions and Predicting Shortages in Resource-Constrained Areas;using Machine Learning to Assess the Impact of Harsh Violent Discipline on Children and Adolescents in Lowand Middle-Income Countries: A Comparative Analysis Focusing on Its Relationship with Disabilities;HealthAIDE: Developing an Audit Framework for AI-Generated Online Health Information;enhancing Diagnostic Accuracy of Drug-Resistant Tuberculosis on Chest X-Rays Using Data-Efficient Image Transformers;CFC Annotator: A Cluster-Focused Combination Algorithm for Annotating Electronic Health Records by Referencing Interface Terminology;novel Approach to De-Identify Relational Healthcare Databases at Rest: A De-Identification of Key Data Approach;enhancing Fracture Aftercare through a Human-Centered Mobile App Design;challenges of Generalizing Machine Learning Models in Healthcare.
Unsupervised domain adaptation gains remarkable progress in real visual tasks by leveraging the learned knowledge from labeled source domain to solve a similar task from unlabeled target domain by adopting pre-trained...
详细信息
the proceedings contain 256 papers. the special focus in this conference is on Biomedical engineering Systems and Technologies. the topics include: Cross-Domain Transfer Learning for Domain Adaptation in Autism Spectr...
the proceedings contain 256 papers. the special focus in this conference is on Biomedical engineering Systems and Technologies. the topics include: Cross-Domain Transfer Learning for Domain Adaptation in Autism Spectrum Disorder Diagnosis;an Interpretable Machine Learning Model for Meningioma Grade Prediction;approximation of Inertial Measurement Unit Data to Time Series Kinematic Data through Correlation Analysis and Machine Learning;Privacy-Preserving Mortality Prediction in ICUs Using Federated Learning;trends in Drug Prescriptions in the Outpatient Physician Sector in a German Federal State from 2014 to 2023 Using Morbidity Related Groups, Correlations and Partial Correlations;quality Clustering for Reducing the Search Space for Mobile Stroke Unit Allocation;leveraging Cross-Verification to Enhance Zero-Shot Prompting for Care Document Data Extraction;Integrating Gait and Clinical Data with Explainable Artificial Intelligence for Parkinson’s Prediction: the EDAM System;Unveiling Breast Cancer Causes through knowledge Graph Analysis and BioBERT-Based Factuality Prediction;optimizing Blood Transfusions and Predicting Shortages in Resource-Constrained Areas;using Machine Learning to Assess the Impact of Harsh Violent Discipline on Children and Adolescents in Lowand Middle-Income Countries: A Comparative Analysis Focusing on Its Relationship with Disabilities;HealthAIDE: Developing an Audit Framework for AI-Generated Online Health Information;enhancing Diagnostic Accuracy of Drug-Resistant Tuberculosis on Chest X-Rays Using Data-Efficient Image Transformers;CFC Annotator: A Cluster-Focused Combination Algorithm for Annotating Electronic Health Records by Referencing Interface Terminology;novel Approach to De-Identify Relational Healthcare Databases at Rest: A De-Identification of Key Data Approach;enhancing Fracture Aftercare through a Human-Centered Mobile App Design;challenges of Generalizing Machine Learning Models in Healthcare.
Nowadays, agile methodologies are being increasingly adopted in the software development industry, replacing traditional methodologies. In this way, softwareengineering courses have been following the industry, and a...
详细信息
ISBN:
(纸本)9798400717819
Nowadays, agile methodologies are being increasingly adopted in the software development industry, replacing traditional methodologies. In this way, softwareengineering courses have been following the industry, and are therefore increasingly teaching students to follow agile methodologies and practices rather than traditional ones. this paper describes and analyzes this transition in a software Project Management course at a higher-education institution. this experiment took place over two academic years, with Waterfall being used in the first year and Scrum in the second. the Learning Outcomes in both years are the same: to gain competences in managing software projects in small teams;but the steps to reach these competences changed according to the current trends in the area. the results obtained by the students show that, by following Scrum, the students demonstrated being more capable of developing software in teams, focused on the clients, and acquired more knowledge in fundamental areas of software development.
暂无评论