检索结果-内蒙古大学图书馆

DiverseNet: Decision Diversified Semi-supervised Semantic Segmentation Networks for Remote Sensing Imagery

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 14258-14271页

作者： Ma, Wanli Karakus, Oktay Rosin, Paul L. Cardiff University School of Computer Science and Informatics Cardiff United Kingdom University of Cambridge Department of Engineering Cambridge United Kingdom

Semi-supervised learning (SSL) aims to help reduce the cost of the manual labelling process by leveraging a substantial pool of unlabelled data alongside a limited set of labelled data during the training phase. Since pixel-level manual labelling in large-scale remote sensing imagery is expensive and time consuming, semi-supervised learning has become a widely used solution to deal with this. However, the majority of existing SSL frameworks, especially various teacher-student frameworks, are too bulky to run efficiently on a GPU with limited memory. There is still a lack of lightweight SSL frameworks and efficient perturbation methods to promote the diversity of training samples and enhance the precision of pseudo labels during training. In order to fill this gap, we proposed a simple, lightweight, and efficient SSL architecture named DiverseHead, which promotes the utilisation of multiple decision heads instead of multiple whole networks. Another limitation of most existing SSL frameworks is the insufficient diversity of pseudo labels, as they rely on the same network architecture and fail to explore different structures for generating pseudo labels. To solve this issue, we propose DiverseModel to explore and analyse different networks in parallel for SSL to increase the diversity of pseudo labels. The two proposed methods, namely DiverseHead and DiverseModel, both achieve competitive semantic segmentation performance in four widely used remote sensing imagery datasets compared to state-of-the-art semi-supervised learning methods. Meanwhile, the proposed lightweight DiverseHead architecture can be easily applied to various state-of-the-art SSL methods while further improving their performance. The code is available at Here. © 2008-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Towards LLM-Generated Affective Phrase Sets for Text Entry Evaluation 24

Towards LLM-Generated Affective Phrase Sets for Text Entry E...

引用

23rd International Conference on Mobile and Ubiquitous Multimedia, MUM 2024

作者： Komninos, Andreas Simou, Ioulia Fotopoulos, Angelos Michanetzi, Eleftheria Lito Xenos, Michalis Computer Engineering & Informatics Department University of Patras Rio Greece

ISBN: (纸本)9798400712838

We present a novel method to generate affective phrase sets using Large Language Models (LLMs), for use in text entry study transcription tasks. These phrases could be used to adjust participant emotions, and enable performance assessment under various affective conditions. We found that the affective phrases impacted participants' valence, but not arousal or text entry performance. Though partially succesful, our work provides a foundation for exploring the relationship between emotion and text entry performance in lab studies. © 2024 Copyright held by the owner/author(s).

关键词： Text Entry

来源：评论

学校读者我要写书评

暂无评论

Personalized User Experience and Engagement in Interactive Mobile Augmented Reality Software Through Fuzzy Weights 19th

Personalized User Experience and Engagement in Interactive M...

引用

19th International Conference on Web Information Systems and Technologies, WEBIST 2023

作者： Papakostas, Christos Troussas, Christos Krouska, Akrivi Mylonas, Phivos Sgouropoulou, Cleo Department of Informatics and Computer Engineering University of West Attica Egaleo Greece

ISBN: (纸本)9783031896200

The integration of digital content into the real world through mobile augmented reality (AR) systems presents captivating prospects, yet user engagement and experience within such environments are largely contingent upon individual computer skills, which can vary significantly among users, thereby influencing their proficiency in utilizing this technology. This paper aims to bridge this gap by examining the impact of users’ computer skills on interactions within mobile AR settings. To this end, a novel fuzzy logic-based model is introduced to evaluate and refine users’ computer skills within the realm of mobile AR systems. By modeling users’ computer aptitude, the system is capable of delivering tailored assistive prompts and feedback. These prompts are strategically aligned with established fuzzy weights, thereby augmenting users’ interactions within the mobile AR sphere. The proposed methodology is seamlessly integrated into a personalized mobile AR platform designed for spatial ability training. Evaluation findings reveal promising outcomes, underscoring the substantial potential of fuzzy logic modeling to elevate user experiences and propel advancements in mobile augmented reality technology. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Teaching

来源：评论

学校读者我要写书评

暂无评论

Partition Strategies for Vertex-Centric Historical Graph Systems 25

Partition Strategies for Vertex-Centric Historical Graph Sys...

引用

40th Annual ACM Symposium on Applied Computing, SAC 2025

作者： Spitalas, Alexandros Tsolas, Georgios Tsichlas, Kostas Computer Engineering and Informatics Department University of Patras Rio Achaia Greece

ISBN: (纸本)9798400706295

This paper explores partitioning strategies for vertex-centric historical graph systems within distributed environments, focusing on efficient data management and query execution. Historical graphs, which capture the dynamic evolution of vertices and edges over time, present unique challenges for storage and computation due to their constantly changing nature. We investigate two partitioning approaches for offline and online environments. Both offline and online algorithms aim to minimize an appropriately defined notion of edge cuts within the historical graph setting;the former algorithm is based on BFS, while the latter is based on a greedy partitioning approach. Both algorithms are compared qualitatively and quantitatively to hash-based methods that are mainly used in this setting. They are evaluated on real-world datasets, using metrics such as weighted edge cut score ratio and load balance ratio. Our experiments reveal that our online method consistently reduces edge cuts with minimal overhead, making it suitable for real-time distributed processing systems for historical graphs. Similarly, the offline algorithm achieves much better results for weighted edge cuts, but it requires severely more computational resources to achieve this performance. Additionally, the study highlights how varying the number of workers and partitioning thresholds impacts system performance across different datasets. All our experiments have been carried out in a simulation environment. The results provide valuable insights into optimizing partition strategies for historical graphs, paving the way for more efficient graph analytics in distributed systems. Copyright © 2025 held by the owner/author(s).

关键词： Parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Comparison of Elbow, Silhouette and DBI Methods for Clustering Nutritional Status of Toddlers Using K-Means Clustering 12

Comparison of Elbow, Silhouette and DBI Methods for Clusteri...

引用

12th International Conference on Cyber and IT Service Management, CITSM 2024

作者： Muharram, Asep Taufik Nalawati, Rizki Elisa Warsuta, Bambang Malik Matin, Iik Muhamad Pradiptyas, Ayres Natanael, Michael Politeknik Negeri Jakarta Informatics and Computer Engineering Depok Indonesia

ISBN: (纸本)9798331528799

Nutritional status is a measure of a person's health condition as seen from the food consumed. The implementation of nutritional monitoring or nutritional surveillance at the Community Health Center begins with data collection and reporting of toddler nutritional status with the Electronic Community-Based Nutrition Recording and Reporting System (E-PPGBM), the recording process must input toddler data from each integrated health post so that the process takes a long time until each integrated health post has measured each toddler. Furthermore, the data is processed and analyzed with Excel to produce a health information report in the form of toddler nutritional status data from each integrated health post which is reported to the health office. In the final process, namely dissemination, namely decision making for follow-up actions to carry out nutritional improvement programs. The nutritional surveillance process still has motivational constraints and performance becomes less than optimal, such as late reporting and coaching because all implementation processes are only carried out by one health center nutrition officer. The health center needs a data clustering system for the nutritional status of toddlers at each integrated health post to facilitate the decision-making process and monitoring of the nutritional health of each integrated health post so that information can be obtained to help nutrition officers prioritize integrated health posts for nutritional improvement service programs in order to achieve improvements in the nutritional health of toddlers at the Abadijaya Health Center. Evaluation of the number of optimal clusters using the Elbow, Silhouette, and Davies Bouldin Index (DBI) methods. The Elbow method obtained 4 optimal clusters, while the Silhouette method obtained 2 optimal clusters and the Davies Bouldin Index (DBI) obtained 2 clusters. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Design of an AI-Assisted Platform that will Facilitate Educators in Creating Minecraft Education Courses 6

Design of an AI-Assisted Platform that will Facilitate Educa...

引用

6th International Workshop on Artificial Intelligence and Education, WAIE 2024

作者： Roumpas, Konstantinos Xenos, Michalis University of Patras Computer Engineering and Informatics Department Patras Greece

ISBN: (纸本)9798350378894

The potential of Education Technology has reached a point where teachers cannot keep up with the latest trends, nor they can spend the time to get accustomed to innovative technologies that might become obsolete in a few years. With the rise in the popularity of Minecraft Education, an educational platform based on the engine of the world-famous video-game: Minecraft, as well as the benefits of gamification in classrooms for all student ages, it seems that an application that would help teachers create educational video games that are run on Minecraft Education, without any code and with artificial intelligence assistance, could be just the thing that they need to update their teaching techniques. We propose the creation of a platform like that, and we present its potential design, after being consulted by the available research that has been done to this day regarding instructional design, user interface and applications that utilize large language models. © 2024 IEEE.

关键词： Gamification

来源：评论

学校读者我要写书评

暂无评论

Use Technology to Help Medical Staff Treat "New Health Problems" Arising Constantly

Use Technology to Help Medical Staff Treat "New Health Probl...

引用

2024 European Federation for Medical informatics Special Topic Conference, EFMI STC 2024

作者： Karanikolas, Nikitas N. Department of Informatics and Computer Engineering University of West Attica Greece

ISBN: (纸本)9781643685540

The development of medical science allows the treatment of more and more health problems that in the past were not a factor of consumption of health resources, because at that time medical science did not have protocols for their treatment. Health problems that are now treatable, hereafter referred as 'new health problems', often affect large population groups and require increased consumption of health resources. It therefore becomes necessary to increase the number of staff providing health services (doctors, nurses, etc.) and other resources. This raises the question: is it feasible to manage the 'new health problems' by the existing medical staff? If not, are there other solutions? Could technology help the existing Medical Staff to sufficiently manage the 'new health problems'? We will examine a pilot system 'Recording and visualizing of outpatient monitoring data with smart mobile phones', which seeks to ensure the competence of existing medical staff in the effective treatment of the ever-increasing volume of transplant patients. © 2024 The Authors.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

IoT-Based Anti-Theft and Violation System with Keyless Ignition for Motorcycle 12

IoT-Based Anti-Theft and Violation System with Keyless Ignit...

引用

12th International Conference on Cyber and IT Service Management, CITSM 2024

作者： Malik Matin, Iik Muhamad Yulianti, Susana Dwi Cahyani, Rifka Nur Sulthanah, Nisrina Tsany Kadarwati, Dinda Marcheta, Noorlela Computer and Informatics Engineering Politeknik Negeri Jakarta Depok Indonesia

ISBN: (纸本)9798331528799

Motorcycles are one of the most flexible means of transportation, especially in urban areas. Motorcycles are prone to theft. This is because access to ignition is easier than other vehicles such as cars. For this reason, a motorcycle security system was developed that has two main systems, namely a tracking system (vehicle tracker) and a keyless ignition system using a LoRa-based Smart SIM (e-driver License) card. The results of functionality and integration tests show that the entire system can be interconnected and work according to the system plan. Performance testing shows that the GPS module performs better when in open space conditions. The average time required by the system to start the vehicle using Smart SIM is 2.83 seconds, start the engine for 6.46 seconds, and turn off the vehicle for 5.18 seconds. The vibration sensor has a presentation error rate of 28% in detecting vibrations. The LoRa module has a greater transmission range in LOS (Line of Sight) conditions. © 2024 IEEE.

关键词： Motorcycles

来源：评论

学校读者我要写书评

暂无评论

Content Creation for Touristic Portals Using ChatGPT 15

Content Creation for Touristic Portals Using ChatGPT

引用

15th International Conference on Information, Intelligence, Systems and Applications, IISA 2024

作者： Lampros, Papantoniou Eleni, Voyiatzaki Ioannis, Hatzilygeroudis University of Patras Computer Engineering and Informatics Dept Patras Greece

ISBN: (纸本)9798350368833

Our research presents the development of an innovative web-based application designed to assist hoteliers in creating and managing multilingual content for their tourist accommodations for tourism portals. To address this need, our application leverages the advanced capabilities of OpenAI's ChatGPT for automated content generation and DeepL's API for high-quality translations, integrated within a user-friendly interface. Through an integrated development process, we leveraged the capabilities of combining natural language processing with artificial intelligence to meet the specific needs of the tourism and hospitality industry. Our findings indicate that the application significantly streamlines the content creation and translation process, offering hoteliers a competitive edge in a globally connected marketplace. By automating these tasks, the system not only saves time and resources but also ensures consistency and quality in accommodation listings across different languages. The application's impact is further evidenced by its adaptability to various accommodation types and the potential for enhancing guest satisfaction through improved information accessibility. The benefits of our application are supported by a medium scale evaluation. © 2024 IEEE.

关键词： Tourism

来源：评论

学校读者我要写书评

暂无评论

A Procedure with MySQL Python for Developing Data Warehouse and Analytics Using Data from a Greek Soft Drinks Company

A Procedure with MySQL Python for Developing Data Warehouse ...

引用

2024 International Conference on Applied Mathematics and computer Science, ICAMCS 2024

作者： Tsakiridis, Sotirios Tsimpiris, Alkiviadis Angeioplastis, Athanasios Papaioannou, Nikolaos Mastorocostas, Paris Varsamis, Dimitrios International Hellenic University Dept. Computer Informatics and Telecommunications Engineering Serres Greece University of West Attica Dept. Informatics and Computer Engineering Egaleo Greece

ISBN: (纸本)9798350363807

The research focuses on the process of creating a data warehouse to meet the decision-making needs of a Greek beverage company. The data cover the period from 2018 to 2022. The developed data warehouse schema follows the star schema, with one fact table and five-dimension tables. Based on this structure, a web application was developed in PHP, which displays reports, charts, and analyses on the company's performance through interactive queries. Additionally, prediction functions tailored to the company's data were developed in Python. The process followed, the queries for creating the data warehouse, as well as the commands for creating analyses and predictions, serve as a useful guide for other companies that wish to upgrade the capabilities of their ERP systems. © 2024 IEEE.

关键词： Data warehouses

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：