检索结果-内蒙古大学图书馆

6th International Conference on AI in Computational Linguistics, ACLing 2024

作者： Ezzat, Zeyad Khalfallah, Ayman Khoriba, Ghadad Center for Informatics Science Information Technology and Computer Science School Nile University Giza Egypt Computer and Systems Department Faculty of Engineering Alexandria University Alexandria Egypt Applied Innovation Centre Egypt Faculty of Computers and Artificial Intelligence Helwan University Cairo Egypt

This paper introduces a novel approach for extending the applicability of pre-trained models to accommodate longer texts. Addressing the inherent limitation of quadratic performance in attention models within transformers, we propose the concept of fused transformers. By integrating new adaptor techniques, we enhance the encoding of lengthy text segments by breaking them into shorter spans. Subsequently, these segments are fused to increase the model's effectiveness in processing extended texts. This fusion mechanism serves to fortify the model's capacity for fine-tuning. Furthermore, we implement a length-based curriculum for expedited training Our experiments yielded 16 Rouge-2 points, representing a doubling of the score achieved by the vanilla fine-tuning method on the newly introduced "Mukhtasar" dataset for summarization. This highlights the effectiveness of managing complex relations among text segments and confirms that our method can outperform conventional training approaches. © 2024 Elsevier B.V.. All rights reserved.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Integration of Large Language Models into Higher Education: A Perspective from Learners 25

Integration of Large Language Models into Higher Education: ...

引用

25th International Symposium on computers in Education, SIIE 2023

作者： Zdravkova, Katerina Dalipi, Fisnik Ahlgren, Fredrik Ss. Cyril and Methodius University Faculty of Computer Science and Engineering Skopje Macedonia Linnaeus University Faculty of Technology Department of Informatics Kalmar Sweden Linnaeus University Faculty of Technology Department of Computer Science and Media Technology Kalmar Sweden

ISBN: (纸本)9798350329315

Large language models (LLMs) are being criticized for copyright infringement, inadvertent bias in training data, a danger to human innovation, the possibility of distributing incorrect or misleading information, and prejudice. Due to their popularity among students, the introduction of many comparable apps, and the inability to resist unfair and fraudulent student usage, their educational use needs to be adapted and harmonized. The incorporation of LLMs should be defined not only by pedagogues and educational institutions, but also by students who will actively utilize them to learn and prepare assignments. In order to find out what students from two universities think and suggest about LLMs use in education, they were asked to give their contribution by answering the survey that was conducted at the beginning of the spring semester of academic 2022/23. Their feedback was quantitatively and qualitatively analyzed, showing in a better light what students think about LLMs and how and why they would use them. Based on the analysis, the authors propose an original strategy for integrating LLMs into education. The proposed approach is also adapted for those students who are not interested in using LLMs and for those who prefer the hybrid mode by combining their own research with LLMs generated recommendations. The authors expect that by implementing the proposed strategy, schools will benefit from a better education in which research, creativity, academic honesty, recognition of false information, and the ability to improve knowledge will prevail. © 2023 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Aspect-Based Subjectivity Analysis Using a BERT-based Approach

Aspect-Based Subjectivity Analysis Using a BERT-based Approa...

引用

2024 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2024

作者： Ng, Hu Chong, Wing Kin Chia, Yu Zhang Vun Yap, Timothy Tzen Goh, Vik Tor Wong, Lai Kuan Teck Tan, Ian Kim Cher, Dong Theng Faculty of Computing and Informatics Multimedia University Cyberjaya Malaysia Faculty of Engineering Multimedia University Cyberjaya Malaysia School of Mathematical & Computer Sciences Heriot-Watt Univeristy Putrajaya Malaysia SIRIM Berhad Shah Alam Malaysia

ISBN: (纸本)9798350353464

Aspect-based subjectivity analysis stands as an important task in natural language processing, seeking to identify the subjectivity of various aspects or features within a text. A new method for aspect-based subjectivity analysis using BERT is introduced in this paper. BERT has demonstrated impressive performance across various NLP tasks, and its capabilities are utilized to accurately ascertain the subjectivity of specific aspects within a given text. The approach involves fine-tuning BERT on a sizable dataset annotated with aspect-level subjectivity labels, enabling the model to grasp the subtleties of aspect-based subjectivity analysis. Extensive experiments on benchmark datasets are conducted to showcase the effectiveness of this approach and compare it with existing methods. The results reveal that this proposed approach surpasses state-of-the-art techniques in aspect-based subjectivity analysis, underscoring the potential of leveraging BERT for such purposes. ©2024 IEEE.

关键词： Aspect-based NLP Subjectivity Analysis

来源：评论

学校读者我要写书评

暂无评论

Enhanced Speech Emotion and Gender Recognition Using Hybrid CNN-LSTM Models 4

Enhanced Speech Emotion and Gender Recognition Using Hybrid ...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Priyadharshini, S. Sundar, G. Naveen Narmadha, D. Thangavel, Senthil Kumar Ajibesin, Adeyemi Abel Kolandapalayam Shanmugam, Selvanayaki Karunya Institute of Technology and Sciences Department of Computer Science and Engineering Coimbatore India Amrita School of Computing Department of Computer Science and Engineering Coimbatore India Cape Peninsula University of Technology Faculty of Informatics and Design Department of Information Technology South Africa Ashland University Department of Mathematics and Computer Science United States

ISBN: (纸本)9798331523923

The current paper proposes a new approach for peripheral speech emotion analysis and gender estimation incorporating the best machine learning architectures such as CNNs and LSTMs. Its correct depiction of emotions and the variation of the gendered speech to indicate manifold signal-to-noise ratio make the approach accommodate the difficulty inherent in recognizing emotions from the recording of voices as well as predicting the gender of a speaker. It is then able to ally with preprocessing and collection of a vast range of different speech samples, feature extraction of pitch, tone and spectral data. These features serve as input to the CNNs and LSTMs: LSTMs work well at temporal regular features and the CNNs work well in feature extraction by location (spatio feature). The approach is tested on the RAVDESS and IEMOCAP datasets and establishes a near perfect match to prior approaches with reference to recognition accuracy of emotion analysis and gender prediction. Thus, applying the identified dual method approach, which could be described by such characteristics as highly accurate data pre-preprocessing, clear indication of the features of interest, a different CNN-LSTM model joint training, it offers a reasonably good platform to gain certain insights into such characteristics of the speech as our experiment demonstrates. This work offers satisfactory configurations on concerns such as human computer relation, customer relation and mental health surveillance to suggest that other applications could be achieved in future. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Improvement of the Control of the Hybrid Photovoltaic System with Storage Battery to Increase the Object's Power Over the Consumption Limit from the Grid 5

Improvement of the Control of the Hybrid Photovoltaic System...

引用

5th IEEE International Conference on Modern Electrical and Energy System, MEES 2023

作者： Shavolkin, Olexander Shvedchykova, Iryna Kolcun, Michal Medved, Dusan Marchenko, Ruslan Kyiv National University of Technologies and Design Department of Computer Engineering and Electromechanics Kyiv Ukraine Technical University of Kosice Faculty of Electrical Engineering and Informatics Košice Slovakia

ISBN: (纸本)9798350359787

The improvement of the control of the hybrid photovoltaic system with a storage battery to increase the power of objects over the limit on consumption from the grid is considered. The aim is to ensure the functioning of the system at the deviation of the load schedule from the calculated one and the generation of photovoltaic battery from the forecast value while maintaining the taken limit of the energy storage depth of discharge. The principle of the control has been improved by setting the value of the added power and forming the state of charge graph of the storage battery. The formation of the storage battery state of charge graph with the correction of the value of the added power according to the deviation of the state of charge from the calculated value with a given step of discreteness in time is proposed. The control algorithm has been changed with the introduction of additional limitations and the exclusion of correction at separate time intervals. Changes are taken into account in the 24-hour model of energy processes in the system. Also, the error of the storage battery state of charge estimation is taken into account. The simulation results confirm the possibility of providing the accepted depth of the discharge value for the calculated load schedule when photovoltaic battery generation is below the forecast value of up to 30%. © 2023 IEEE.

关键词： Electric power supplies to apparatus

来源：评论

学校读者我要写书评

暂无评论

DO DEEP NEURAL NETWORKS UTILIZE THE WEIGHT SPACE EFFICIENTLY?

arXiv

引用

arXiv 2024年

作者： Koyun, Onur Can Töreyin, Behçet Ugur Department of Computer Engineering Faculty of Computer and Informatics Istanbul Technical University Istanbul Turkey Department of Artificial Intelligence and Data Engineering Faculty of Computer and Informatics Istanbul Technical University Istanbul Turkey

Deep learning models like Transformers and Convolutional Neural Networks (CNNs) have revolutionized various domains, but their parameter-intensive nature hampers deployment in resource-constrained settings. In this paper, we introduce a novel concept utilizes column space and row space of weight matrices, which allows for a substantial reduction in model parameters without compromising performance. Leveraging this paradigm, we achieve parameter-efficient deep learning models.. Our approach applies to both Bottleneck and Attention layers, effectively halving the parameters while incurring only minor performance degradation. Extensive experiments conducted on the ImageNet dataset with ViT and ResNet50 demonstrate the effectiveness of our method, showcasing competitive performance when compared to traditional models. This approach not only addresses the pressing demand for parameter efficient deep learning solutions but also holds great promise for practical deployment in real-world scenarios. © 2024, CC BY-NC-ND.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

An Augmented Reality Environment for Testing Cockpit Display Systems 7th

An Augmented Reality Environment for Testing Cockpit Displa...

引用

7th International Conference on computer-Human Interaction Research and Applications, CHIRA 2023

作者： Potur, Caner İnce, Gökhan MGEO Test and Verification Directorate ASELSAN Inc. Ankara Turkey Faculty of Computer and Informatics Engineering Istanbul Technical University Istanbul Turkey

ISBN: (纸本)9783031493676

In the avionics industry, software test automation is a crucial component, as it plays a significant role in accelerating the development process. Since all automation systems are error-prone, the automation systems that affect flight safety should be supervised by humans as they must be reliable. In this study, an augmented reality based software verification system is designed to observe the visual verification of cockpit screens and detect possible automation errors. In the designed system, an observer can instantly see the test steps and automation results using an augmented reality headset as a hologram on the cockpit screen and can report the errors they observe in test automation. The performance of the proposed system was analyzed qualitatively and quantitatively. The experimental results demonstrated that the use of augmented reality in cockpit display verification systems accelerated the testing process, reduced users’ cognitive load, and improved the system’s usability. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Automation

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Chaotic Electromagnetic Field Optimization Algorithms

引用

International Journal Bioautomation 2024年第4期28卷 245-265页

作者： Roeva, Olympia Zoteva, Dafina Department of Bioinformatics and Mathematical Modelling Institute of Biophysics and Biomedical Engineering Bulgarian Academy of Sciences Acad. G. Bonchev Str. bl. 105 Sofia1113 Bulgaria Department of Computer Informatics Faculty of Mathematics and Informatics Sofia University "St. Kliment Ohridski" 5 James Bourchier Blvd. Sofia1164 Bulgaria

This paper investigates the performance of various Electromagnetic Field Optimization (EFO) algorithms. Chaos maps are proposed to improve the performance of EFO algorithms. Ten chaotic maps are incorporated in EFO – Chebyshev, Circle, Gauss, Iterative, Logistic, Piecewise, Sine, Singer, Sinusoidal and Tent. To compare the performance of the constructed EFO algorithms, a case study of the identification of the model parameters of a cultivation process model is studied. An experimental data set from E. coli BL21(DE3)pPhyt109 fed-batch cultivation process is used. Based on the results of 30 runs of each EFO, some statistical and InterCriteria analyzes are performed. As a result, the best performing EFO algorithms are iterative EFO and tent chaotic map EFO. These algorithms gave the best objective value (best and mean value) and had a good distribution of results. © 2024 by the authors.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

A Process Oriented Integration Model for Smart Health Services

引用

Intelligent Automation & Soft Computing 2023年第2期35卷 1369-1386页

作者： Farzana Kausar Gondal Syed Khuram Shahzad Muhammad Arfan Jaffar Muhammad Waseem Iqbal Department of Computer Science Superior UniversityLahore54600Pakistan Department of Informatics and Systems University of Management and TechnologyLahore54000Pakistan Faculty of Computer Science&IT Superior UniversityLahore54600Pakistan Department of Software Engineering Superior UniversityLahore54600Pakistan

Cities are facing challenges of high rise in population number and con-sequently need to be equipped with latest smart services to provide luxuries of life to its *** integrated solutions are also a need to deal with the social and environmental challenges,caused by increasing ***,the development of smart services’integrated network,within a city,is facing the bar-riers including;less efﬁcient collection and sharing of data,along with inadequate collaboration of software and *** to resolve these issues,this paper recommended a solution for a synchronous functionality in the smart services’integration process through modeling *** this integration modeling solution,atﬁrst,the service participants,processes and tasks of smart services are identiﬁed and then standard illustrations are developed for the better understand-ing of the integrated service group *** process modeling and notation(BPMN)language based models are developed and discussed for a devised case study,to test and experiment i.e.,for remote healthcare from a smart *** research is concluded with the integration process model application for the required data sharing among different service *** outcomes of the modeling are better understanding and attaining maximum automation that can be referenced and replicated.

关键词： Process modeling business process business process integration business process modeling and notation

来源：评论

学校读者我要写书评

暂无评论

Comparison of The Accuracy of Predicting The Number Of Positive Covid-19 Between The Neural Network and LSTM Methods

Comparison of The Accuracy of Predicting The Number Of Posit...

引用

2023 International Conference on informatics, Multimedia, Cyber and Information Systems, ICIMCIS 2023

作者： Riyanto, Eko Amalia, Rifka Dwi Stmik Himsya Informatics Engineering Semarang Indonesia Universitas Pembangunan Nasional Veteran Jakarta Information Systems Faculty of Computer Science Jakarta Indonesia

ISBN: (纸本)9798350307306

The COVID-19 pandemic (Coronavirus) is likely to be one of the most serious global problems in the last year. Countries do not have similar experiences with the spread of the virus and its impact from various fields. Estimating the number of COVID-19 cases in the past can help in decision-making about action and plan for virus prevention. The study aims to provide accurate predictive model in predicting confirmed cases of COVID-19 in Semarang city. In predicting the number of people exposed to Covid-19 positive need to be accurate and accurate methods, in some previous studies the two methods most recommended are neural network and LSTM (Long Short Term Memory) which have high levels of accuracy. Based on research that has been carried out that the results of accurate prediction comparison between RNN and LSTM obtain the smallest MSE result is the method. © 2023 IEEE.

关键词： Forecasting

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：