检索结果-内蒙古大学图书馆

2024 International Conference on applied Mathematics and Computer Science, ICAMCS 2024

作者： Ceco, Polina Hoxha, Elira University of Tirana Faculty of Economy Department of Statistics and Applied Informatics Albania

ISBN: (纸本)9798350363807

Natural Language Processing (NLP) has been widely used to surface the intelligence embedded in textual documents that could not be leveraged through traditional data mining techniques. While NLP models can be shifted with almost no change across domains and languages, training materials such as corpora cannot. Currently, there are no available domain-specific corpora for the training of medical NLP models in the Albanian language. Therefore, in this paper, we introduce the creation of the Medical Dissertations Corpus in Albanian (MeDA1), a 6M words corpus in Albanian in the medical domain. The corpus was built from parsing and cleaning more than 300 dissertations published in the last 10 years by the University of Medicine in Tirana. The corpus can be used in the training or continuous training of NLP models used in medical settings - we intend to use the corpus in a Named Entity Recognition (NER) task for Electronic Health Records (EHR) in the Albanian language. In this paper, we describe in detail the process of PDF parsing and text cleaning that we have conducted for the creation of MeDAl. © 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

Using AI to Improve Accessibility and Inclusivity in Higher Education for Students with Disabilities 21

Using AI to Improve Accessibility and Inclusivity in Higher ...

引用

21st International Conference on Information Technology Based Higher Education and Training, ITHET 2024

作者： Mitre, Xhulio Zeneli, Muhamet University of Tirana Faculty of Economics Department of Statistics and Applied Informatics Tirana Albania

ISBN: (纸本)9798331516635

Around 1.3 billion, or more than 15% of the global population, suffer from a significant form of disability, with students and researchers in this category being underrepresented in higher education. This paper examines the use of Interactive Artificial Intelligence solutions to improve the inclusivity and accessibility of students with disabilities in higher education. The paper addresses the failure of traditional education methods to accommodate the diverse needs of people with disabilities. Through an exploratory literature review, several AI-driven solutions, like assistive technologies, adaptive-learning systems, and generative AI chatbots and virtual assistants, have been reviewed to understand how AI could be leveraged to create an inclusive learning environment. Key results show AI-driven solutions' potential to transform the learning process of people with disabilities by creating personalized learning paths, increasing access to educational resources, and supporting real-time communication. Additionally, the research highlights ethical concerns about the integration of AI in education, emphasizing the need for the participation of disabled individuals in the developmentprocess. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing University Students' Activities with Interactive Immediate Feedback Using Customized LLMs 12

Enhancing University Students' Activities with Interactive I...

引用

12th International Scientific Conference on Computer Science, COMSCI 2024

作者： Andonov, Venko University of National and World Economy Faculty of Applied Informatics and Statistics Department of Informatics 8 Kliment Ohridski blvd. Sofia1000 Bulgaria

ISBN: (纸本)9798350392166

One of the significant issues for the students is getting timely feedback on their work so that they can understand their mistakes and improve their knowledge iteratively. There have been significant results in this area through automated evaluation, but they require a strongly defined structure of the expected results, which has its limitations. We propose using customized large language models as the core of an assignment evaluation software architecture that is used for automated evaluation and feedback generation on a variety of tasks in an introductory Java programming course. The model is implemented and tested, achieving significantly better results than using general-purpose models. © 2024 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Opportunities to Improve the Cybersecurity of Software Applications 12

Opportunities to Improve the Cybersecurity of Software Appli...

引用

12th International Scientific Conference on Computer Science, COMSCI 2024

作者： Naskova, Gabriela Dimitrova University of National and World Economy Faculty of Applied Informatics and Statistics Department of Informatics 19 8th December blvd. Sofia1700 Bulgaria

ISBN: (纸本)9798350392166

This study focuses on issues related to the cybersecurity of software applications. A review of key literature sources on technological security measures for information systems and web platforms is presented. Some measures in various stages of the software development process are critically examined and analyzed. In support of the presented concepts, an online survey was conducted. The data from the survey were analyzed, and based on this, conclusions and recommendations for software engineers and system programmers were formulated. © 2024 IEEE.

关键词： cybersecurity programming software applications technological measures web applications

来源：评论

学校读者我要写书评

暂无评论

Moss spores: overlooked airborne bioparticles in an urban environment

引用

Environmental Science and Pollution Research 2024年第47期31卷 58010-58020页

作者： Ščevková, Jana Tropeková, Mária Dušička, Jozef Štefániková, Natália Žilka, Matúš Zahradníková, Eva Kováč, Jozef Mišíková, Katarína Faculty of Natural Sciences Department of Botany Comenius University Révová 39 Bratislava811 02 Slovakia Faculty of Mathematics Physics and Informatics Department of Applied Mathematics and Statistics Comenius University Mlynská dolina Bratislava842 48 Slovakia

Moss spores are present in aerobiological samples, but their low representation, lack of known allergenic properties, and difficult identification have led to their being overlooked by aerobiologists so far. The data about their presence in the atmosphere and the factors that influence them are, however, important from the biodiversity conservation point of view, since they give us information about their ability to spread to new habitats. In this pilot study, we analysed their presence in the atmosphere of Bratislava city, Slovakia (2018–2023), using Burkard volumetric sampler, and determined the most significant factors influencing its temporal distribution. The size category of 13–18 µm was the most represented in the samples. Environmental factors influenced the daily spore concentrations and the characteristics of the whole spore season. The start of the Main Spore Season (MSS) depended mainly on the temperatures in November–January, initiating earlier growth of sporophytes in the following year, while the intensity of the MSS was influenced by high humidity in April, stimulating the formation of spores in sporangia. The daily concentration of airborne moss spores was mostly influenced by the actual temperature and wind speed, promoting the release and dispersal of spores, and precipitation lowering their levels due to the "wash-out" effect, although no intradiurnal pattern was observed. More data from other locations is needed to determine the role of atmospheric spore transport for the conservation of moss species facing anthropogenic climatic change. © The Author(s) 2024.

关键词： Biotic

来源：评论

学校读者我要写书评

暂无评论

Impact of Gingival Phenotype on the Periodontal Disease

引用

Folia Medica 2023年第3期65卷 468-475页

作者： Pashova-Tasseva, Zdravka Mlachkova, Antoaneta Tosheva, Ekaterina Department of Periodontology Faculty of Dental Medicine Medical University of Sofia Sofia Bulgaria Faculty of Applied Informatics and Statistics University of National and World Economy Sofia Bulgaria

Aim: The aims of the present study were to explore the relations between the gingival phenotype (GP) and the periodontal health status and find the prevalence of a specific gingival phenotype in a small Bulgarian population. Materials and methods: We recruited 50 patients attending our dental practice with different periodontal diagnoses. A detailed periodontal status was taken to assess the diagnosis of each participant. Several clinical methods for evaluation of the gingival thickness and keratinized tissue width, including the TRAN method, transgingival probing, and direct measurement, were used. The data were summarized and analyzed statistically. Results: We found a significant prevalence of the thick gingival phenotype, in particular the thick flat type. The patients with periodontitis had a higher distribution of the thick gingival phenotype, while in those with gingivitis, the thin scalloped gingival phenotype was noted. In regards to the gingival thickness (GT), 36 participants were found to have GT >1 mm, and the remaining 14 had GT ≤1 mm. Statistically significant differences were found in the keratinized tissue width and the width of attached gingiva in the different gingival phenotypes. No significant differences were found in the age and sex of participants. Conclusions: We found a significant prevalence of the thick (with a mild prevalence of thick flat to thick scalloped) versus thin gingival phenotype in the studied sample. The highest relative proportion of patients with periodontitis was among the subsample of individuals with thick flat gingival phenotype. Regarding gingivitis, the highest proportion was in the thin scalloped phenotype subsample - 42.9%. The highest prevalence of periodontal health was among the individuals with thick scalloped GP (50%), followed by the thin scalloped GP (35.7%). © 2023 Authors. All rights reserved.

关键词： gingival phenotype gingival thickness keratinized tissue width periodontal phenotype width of attached gingiva

来源：评论

学校读者我要写书评

暂无评论

Analysis of Cybersecurity Competencies as A Factor for Organizations' Cyber Defense 12

Analysis of Cybersecurity Competencies as A Factor for Organ...

引用

12th International Scientific Conference on Computer Science, COMSCI 2024

作者： Kirilova, Katya Emilova University of National and World Economy Faculty of Applied Informatics and Statistics Department of Pedagogy 19 8th December blvd. Sofia1700 Bulgaria

ISBN: (纸本)9798350392166

The article examines contemporary cyber challenges in organizations. A theoretical review of sources on the issues of cyber security competences and the European regulatory framework has been carried out. A developed methodology for research and evaluation of some of the important competencies in the field of cyber security is proposed. A specialized empirical study has also been developed. Its data has been summarized and analyzed, and the results show a growing need for training to improve users' knowledge, skills and competencies. © 2024 IEEE.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

Forecasting LEK/EUR Exchange Rate: A Comparative Deep Learning Study 5

Forecasting LEK/EUR Exchange Rate: A Comparative Deep Learni...

引用

5th International Conference on Recent Trends and Applications in Computer Science and Information Technology, RTA-CSIT 2023

作者： Vika, Blerina Çollaku, Denada University of Tirana Faculty of Economy Department of Statistics and Applied Informatics Arben Broci Tirana Albania University of Tirana Faculty of Natural Sciences Department of Informatics Bulevardi Zogu I 25/1 Tirana Albania

This study contributes to the deep learning literature by investigating the applicability of different DL models to forecasting monthly LEK/EUR exchange rate. To demonstrate the effectiveness for exchange rate forecasting of this models we examine the performance of several deep learning techniques (DFNN, LSTM and 1D-CNN). For each architecture, we have used different configuration and diverse techniques to avoid overfitting of the models. The accuracy evaluation of each model was based on the out-of-sample prediction for different horizon (3, 6 and 12 months ahead), by analyzing the model estimation for pre- and post- Covid pandemic. The comparison results show that the LSTM model with two hidden layers stands out as the best prediction model in the run-up to forecast three and six months ahead, followed by the three-layered 1DCNN model. However, they change places in the race for the 12 months horizon as the 1DCNN-3L becomes the first-best predicting model while leaving the LSTM-2L rank in second. These results demonstrate the potential of deep learning techniques, and also, they emphasize the importance of well configuring, implementing and selecting the different topologies. © 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Length-Minimizing LED Trees

引用

Operations Research Forum 2025年第1期6卷 1-38页

作者： Sarkociová Remešíková, Mariana Sarkoci, Peter Trnovská, Mária Department of Mathematics and Constructive Geometry Faculty of Civil Engineering Slovak University of Technology in Bratislava Radlinského 11 Bratislava 81005 Slovakia Department of Applied Mathematics and Statistics Faculty of Mathematics Physics and Informatics Mlynská Dolina F1 Bratislava 84248 Slovakia

In this paper, we introduce a previously not studied type of Euclidean tree called LED (Leaves of Equal Depth) tree. LED trees can be used, for example, in computational phylogeny, since they are a natural representative of the time evolution of a set of species in a feature space. This work is focused on LED trees that are length minimizers for a given set of leaves and a given isomorphism type. The underlying minimization problem can be seen as a variant of the classical Euclidean Steiner tree problem. Even though it has a convex objective function, it is rather non-trivial, since it has a non-convex feasible set. The main contribution of this paper is that we prove the uniqueness of a stationary point of the length function on the feasible set. Moreover, we prove several geometrical characteristics of the length minimizers that are analogous to the properties of Steiner minimal trees. We also explore some geometrical and topological properties of the feasible set. At the end, to demonstrate the applicability of our theoretical results, we show an example of an application in historical linguistics. © The Author(s) 2025.

关键词： Computational phylogeny Euclidean graph Euclidean tree Language evolution LED tree Length-minimizing tree Phylogenetic tree Steiner tree

来源：评论

学校读者我要写书评

暂无评论

MeDAl – The Creation of Medical Dissertations Corpus in Albanian for Medical NLP

MeDAl – The Creation of Medical Dissertations Corpus in Alb...

引用

International Conference on applied Mathematics & Computer Science (ICAMCS)

作者： Polina Çeço Elira Hoxha Department of Statistics and Applied Informatics Faculty of Economy University of Tirana Albania

ISBN: (数字)9798350363807

ISBN: (纸本)9798350363814

关键词： Training Vocabulary Pipelines Text summarization Named entity recognition Portable document format Cleaning Fuels Electronic medical records Surface treatment

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：