检索结果-内蒙古大学图书馆

Electrical, electronic and Computing engineering (IcETRAN), International Conference on

作者： Nikola Đorđević Suzana Stojković Department of computer science Faculty of electronic engineering University of Niš Niš Serbia

ISBN: (数字)9798350386998

ISBN: (纸本)9798350387001

Fine-tuning of large language models is often demanding in terms of computational resources and memory. Consequently, there is a need to explore new methods that can effectively fine-tune these models without compromising performance. In this paper, we compared traditional and parameter-efficient fine-tuning of large language model XLM-RoBERTa, across both high-resourced languages like English and low-resourced languages such as Serbian. The model is fine-tuned for sentiment analysis when it is conducted using text classification. The experiments have shown that parameter-efficient fine-tuning can produce results that are on par with traditional fine-tuning, both for the English and the Serbian language. Furthermore, parameter efficient fine-tuning approach significantly reduces the computational time and memory requirements associated with model training, thereby offering a promising avenue for resource-efficient large language models fine-tuning. In experiments with the English language, we achieved accuracy using the PEFT approach that was only 3% lower than the accuracy obtained through the traditional approach, while reducing training time by 35% and memory requirements by 42%. In experiments involving the Serbian language, we achieved accuracy using the PEFT approach that was only 1% lower than the accuracy obtained through the traditional approach, while reducing training time by 42% and memory requirements by 69%.

关键词： Training Sentiment analysis Analytical models Accuracy Quantization (signal) Reviews Computational modeling Large language models Memory management Text categorization

来源：评论

学校读者我要写书评

暂无评论

High accuracy, fully reconfigurable DAQ system for sensing temperature 33

High accuracy, fully reconfigurable DAQ system for sensing t...

引用

33rd International Scientific Conference electronics, ET 2024

作者： Gatev, Vasil D. Mollov, Valentin S. Petrov, Panayot L. Manoeva, Mariana D. Technical University of Sofia Faculty of Computer Systems and Technologies Department of Computer Systems 8 Kliment Ohridski blvd. Sofia1000 Bulgaria Technical University of Sofia Faculty of Electronic Engineering and Technologies Department of Electronics 8 Kliment Ohridski blvd. Sofia1000 Bulgaria

ISBN: (纸本)9798350376449

The FPGA configurable integrated circuits (ICs) are a valid alternative to the common microcontroller solutions for application-specific (ASIC) embedded applications. The FPGAs are applied in the prototype step, before the ASIC synthesis, for end-user devices targeted to real-time applications. Nowadays IoT solutions with post-processing on cloud-based server technology, makes the dedicated DAQ device an important component in the system. The FPGAs are the appropriate basis for such device, where the data acquisition of analog values and its transfer to digital representation must be achieved with high-speed, high-resolution, low power consumption and effective optimization of the IC resources. © 2024 IEEE.

关键词： Application specific integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Development of an Open-Source and Low-Cost Robotic Arm based on STM32 Microcontroller for Promoting STEM Education

Development of an Open-Source and Low-Cost Robotic Arm based...

引用

2023 IEEE Symposium on computers and Informatics, ISCI 2023

作者： Asri, Fatin Atirah Mohd Ayob, Mohammad Afif Rohaziat, Nurasyeera Rahman, Hisyam Abdul Rahman, Khairul Azlan A. Ayob, Mohd Erdi Universiti Tun Hussein Onn Malaysia Faculty of Electrical and Electronic Engineering Batu Pahat Parit Raja86400 Malaysia Uow Malaysia Kdu University College School of Engineering Selangor Shah Alam40150 Malaysia Faculty of Electronics and Computer Technology and Engineering Melaka Durian Tunggal76100 Malaysia

ISBN: (纸本)9798350302363

As a key component of Industry 4.0, robotics is acquiring traction in classrooms around the world. Robotics initiatives may stimulate students' interest in STEM (Science, Technology, engineering, and Mathematics) subjects. Therefore, it is essential to translate theoretical information into tangible prototypes and study robotics using physical models. Most schools cannot afford to provide students with sufficient opportunities to work with robots due to the expense of robotics. This study aims to promote STEM education by describing the development of a free, open-source robot joint with three degrees of freedom (DOF). The control algorithm for a 3D-printed MK2 robot arm was programmed with an STM32 microcontroller. The robot has a servo-powered arm that has a gripper on the end-effector. A joystick can be used to control the movement of any mechanical joint. The forward kinematic has also been constructed and validated in a simulation-based environment in MATLAB and SolidWorks. The developed kinematics algorithm showed precise Denavit-Hartenberg (DH) characteristics with high accuracy with the worst-case scenario for the kinematic computation error contributes to only 0.26% or 0.0702mm of the end-effector position. © 2023 IEEE.

关键词： Kinematics

来源：评论

学校读者我要写书评

暂无评论

Enhancing Persian Word Sense Disambiguation with Large Language Models Techniques and Applications

Enhancing Persian Word Sense Disambiguation with Large Langu...

引用

International eConference on computer and Knowledge engineering (ICCKE)

作者： Fatemeh Zahra Arshia Saeedeh Sadat Sadidpour Faculty of Electronic & Computer Engineering Malek Ashtar University of Technology Tehran Iran

ISBN: (数字)9798331511272

ISBN: (纸本)9798331511289

WSD means the task of word sense disambiguation, which is a very important task in NLP. It assigns not only the meaningful word to the source text but also the proper meaning of the word according to the context. Hence, it is key to the proper accomplishment of NLP in Persian-a language rich in morphology and great polysemy. The recent improvements in LMs have greatly advanced the capabilities of NLP, opening further improvement avenues in WSD performance. This paper presents the integration of LLMs for improving WSD in Persian, considering the linguistic challenges related to this language. In this study, we consider four models of the Persian language: FaBERT, AriaBERT, GPT-2 Persian, and PersianMind-v.10 We use the supervised fine-tuning method on the SBU-WSD-Corpus. Our methodology will consist of preprocessing the Persian WSD corpus, then fine-tuning the models with the mentioned corpus, and measuring their performance. Results indicate that methods using LLMs significantly improve WSD accuracy against traditional methods, with FaBERT achieving the best accuracy. We have further expounded on their real-life applications, such as sentiment analysis, to show the consequential effect of this advancement on general NLP tasks. The study is concluded with some insights into future research directions, underlining the potential role that LLMs can play in further transforming WSD and related fields.

关键词： Knowledge engineering Sentiment analysis Accuracy Computational modeling Large language models Morphology Linguistics

来源：评论

学校读者我要写书评

暂无评论

Determining category metadata in open data portals – an approach based on Formal Concept Analysis

Determining category metadata in open data portals – an app...

引用

Telecommunications Forum (TELFOR)

作者： Milena Frtunić Gligorijević Miloš Bogdanović Leonid Stoimenov Department for Computer Science Faculty of Electronic Engineering University of Niš Niš Serbia

ISBN: (数字)9798350391060

ISBN: (纸本)9798350391077

The efforts for data transparency and open government initiatives have resulted in a large amount of data being published on open data portals. These portals are organized to enhance published data accessibility by providing search mechanisms that utilize metadata such as category, tags, format, and publisher. However, due to the metadata incompleteness, such as missing category information, datasets are harder to find. Therefore, within this paper, we address the issue of missing category information by proposing an approach for categorizing datasets using keywords and their usage within categories. The presented approach is based on the Formal Concept Analysis method, which is used to create a hierarchical order of keywords grounded on their usage within categories on one open data portal. Further, we present the evaluation of the presented method using data from the Canadian open data portal.

关键词： Formal concept analysis Knowledge based systems Government Lattices Metadata Market research Iterative algorithms Telecommunications Portals Open data

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Control System Based on Spatial-Temporal Graph Convolutional and Disentangled Baseline-Volatility Prediction of Bellows Temperature for Iron Ore Sintering Process

SSRN

引用

SSRN 2024年

作者： Chi, Zhengwei Chen, Xiaoxia Xia, Hanzhong Liu, Chengshuo Wang, Zhen Faculty of Electronic Engineering and Computer Science Ningbo University Ningbo315211 China

The temperature inside the furnace plays a crucial role in determining the quality of sinter ore of iron ore sintering process. In practical production, the temperature of the bellows, which is directly connected to the bed layer, indirectly reflects the temperature inside the furnace. A stable bellows temperature, especially the temperature at the location of Burn-Through Point, is essential for reducing gas emissions, improving carbon efficiency, and enhancing the quality of sintered ore. This paper proposed a Model Reference Adaptive Control (MRAC) intelligent temperature control system based on Spatial-Temporal Graph Convolutional and Disentangled Baseline-Volatility (STGCDBV) prediction. STGCDBV network consists of three parallel modules (Adaptive Graph Convolution Network (AGCN), Baseline and Volatility Disentangler (BVD) and a residual link) and Temporal-Nodal Encoder-Decoder (TNED) module. AGCN constructs a graph based on the characteristics of the bellows temperature, effectively combining the static spatial information of the bellows with the dynamic temperature information. BVD models the nonlinear trend information of the sintering process. TNED combines the advantageous information from the parallel modules through cross encoding and decoding. MRAC system is employed as the controller for the sintering gas flow rate. It adopts a control approach based on a temperature reference model and parameter iteration to regulate the gas flow rate. Additionally, extensive experiments are conducted using original time series data from a steel plant. Furthermore, a comparison is made between the results of the front and back controls, demonstrating that STGCDBV-MRAC can stabilize temperature fluctuations and exhibit excellent control performance. © 2024, The Authors. All rights reserved.

关键词： Sintering

来源：评论

学校读者我要写书评

暂无评论

An Approach for Metadata Enrichment on Open Data Portals

An Approach for Metadata Enrichment on Open Data Portals

引用

Electrical, electronic and Computing engineering (IcETRAN), International Conference on

作者： Milena Frtunić Gligorijević Miloš Bogdanović Leonid Stoimenov Department for Computer Science Faculty of Electronic Engineering University of Niš Niš Serbia

ISBN: (数字)9798350386998

ISBN: (纸本)9798350387001

Open data initiatives have resulted in a large amount of data being published on open data portals. In order to make published data more accessible these portals provide search mechanisms based on metadata like category, tags, format, and publisher. However, due to the lack of metadata information, like category values, search results are incomplete, and data is harder to be found. Therefore, this paper presents an approach for metadata enrichment on open data portals. In particular, the presented approach is focused on suggesting categories one dataset should belong to, based on the combination of tags that is used for its description. This approach relies on the tags usage for describing datasets within categories and the Formal Concept Analysis method for creating a hierarchical order of tags within categories. The presented proposal has been evaluated using data from Ireland's open data portal.

关键词： Formal concept analysis Data preprocessing Lattices Metadata Tagging Complexity theory Proposals

来源：评论

学校读者我要写书评

暂无评论

Quick-MIMIC:A Multimodal Data Extraction Pipeline for MIMIC with Parallelization

引用

Big Data Mining and Analytics 2024年第4期7卷 1333-1346页

作者： Yutao Dou Wei Li Yangtao Zheng Xiaojun Yao Huanxiang Liu Albert Y.Zomaya Shaoliang Peng College of Computer Science and Electronic Engineering Hunan UniversityChangsha 410082Chinaand also with Centre for Distributed and High Performance ComputingSchool of Computer ScienceThe University of SydneyDarlingtonNSW 2008Australia College of Computer Science and Electronic Engineering Hunan UniversityChangsha 410082China Centre for Distributed and High Performance Computing School of Computer ScienceThe University of SydneyDarlingtonNSW 2008Australia Faculty of Applied Sciences Macao Polytechnic UniversityMacao 999078China

Medical big data with artificial intelligence are vital in advancing digital ***,the opaque and non-standardised nature embedded in most medical data extraction is prone to batch effects and has become a significant obstacle to reproducing previous *** paper aims to develop an easy-to-use time-series multimodal data extraction pipeline,Quick-MIMIC,for standardised data extraction from MIMIC *** method can fully integrate different data structures into a time-series table,including structured,semi-structured,and unstructured *** also introduce two additional modules to Quick-MIMIC,a pipeline parallelization method and data analysis methods,for reducing the data extraction time and presenting the characteristics of the extracted data *** extensive experimental results show that our pipeline can efficiently extract the needed data from the MIMIC dataset and convert it into the correct format for further analytic tasks.

关键词： MIMIC dataset data extraction pipeline data integration

来源：评论

学校读者我要写书评

暂无评论

Efficient Earth Observation Data Visualization Using Dual Orthogonal Equidistant Cylindrical Projection

Efficient Earth Observation Data Visualization Using Dual Or...

引用

Electrical, electronic and Computing engineering (IcETRAN), International Conference on

作者： Aleksandar Dimitrijević Department of Computer Science Faculty of Electronic Engineering University of Niš Niš Serbia

ISBN: (数字)9798350386998

ISBN: (纸本)9798350387001

Earth observation (EO) data have seen a constant surge in volume, necessitating efficient storage, retrieval, and sharing mechanisms. Cartographic projections play a vital role in transforming spheroidal surface data into two-dimensional representations, albeit with distortions. The Dual Orthogonal Equidistant Cylindrical (DOEC) projection offers a balance between simplicity and distortion, making it a promising choice for EO data storage and visualization. This paper explores various methods for raster geodata visualization in DOEC projection, addressing challenges of partitioning, distortion, and visualization techniques. Through experimental comparisons, the paper demonstrates the efficiency of these methods, highlighting their applicability in EO data processing and visualization systems.

关键词： Earth Prevention and mitigation Data visualization Memory Distortion Data processing Real-time systems

来源：评论

学校读者我要写书评

暂无评论

AI-enhanced cognitive behavioral therapy with novel ontological and pattern-based thought segmentation

引用

International Journal of Information Technology (Singapore) 2025年 1-8页

作者： Vidanage, Kaneeka Maduranga, M.W.P. Department of Computer Science Faculty of Computing General Sir John Kotelawala Defense University Kandawala Road 10390 Dehiwala-Mount Lavinia Sri Lanka Department of Electrical and Electronic Engineering University of Sri Jayewardenepura Nugegoda 10500 Sri Lanka

Use of Artificial Intelligence (AI) has contributed to improving the quality standards of the healthcare solutions provided for the patients. Therefore, healthcare informatics has been recognized as a very demanding discipline which is on constant evolution. Thus, the utilization of AI advancements to uplift the treatment strategies are mainly confined to provide solutions linked with treating physical ailments. Application of AI to improve psychotherapeutic treatments are very naïve. However, psychological well-being is also a vital requirement ensuring the physical well-being of an individual. Cognitive Behavioral Therapy (CBT) is recognized as the world's most acceptable psychotherapeutic treatment tool. Consequently, with the involvement of the psychological consultants and referring the CBT manuals, specialized ontological structure has been orchestrated to assess the human “thoughts” as per the CBT guidelines. The designed ontological structure has been accompanied with a specialized knowledge elicitation algorithm designed to reason the knowledge constructs embedded in the ontology against the stakeholder`s input. Operational effectiveness of this entire pipeline has been assessed via both quantitative and qualitative measurements under the observation of the consultants. The entire prototype designed yielded an overall accuracy of 82%. This entire pipeline compiled can be considered as a novel contribution enriching the expansion of the application of AI into the psychotherapeutic domain as well. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： AI CBT Ontology Psychotherapy RegEX

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：