检索结果-内蒙古大学图书馆

Influence of Feature Extraction Methods using Neighborhood Component Analysis (NCA) to Improve Classification Performance for Hepatitis C Disease 9

Influence of Feature Extraction Methods using Neighborhood C...

引用

9th International Conference on Informatics and Computing, ICIC 2024

作者： Setiawan, Christiella Abinosy Anggreainy, Maria Susan Fitriawati, Nora Lucky, Raymasterio Vera Sabrina, Renatha Dewi Matthew, Steven Wijayanti, Vierdaria Bina Nusantara University School of Computer Science Computer Science Department Jakarta 11480 Indonesia Bina Nusantara University Binus Graduate Program Computer Science Department Jakarta 11530 Indonesia Bina Nusantara University Binus Online Learning Information Systems Department Jakarta 11530 Indonesia

ISBN: (纸本)9798331517601

Research in the development of Hepatitis C disease prediction is increasingly developing, especially using machine learning models which is able to make predictions quickly and accurately. In this study, a comparison of several classification methods was carried out by also applying feature reduction, NCA. In this study, a comparison of performance was carried out if the data was entered into the NCA feature extraction method with KNearest Neighborhood (KNN) and Support Vector Machine (SVM) and a comparison of performance if the data did not use the NCA feature extraction method. The performance comparison metrics used in the study were accuracy, sensitivity, specification, Matthews Correlation Coefficient (MCC), and Kappa value. The highest accuracy (99.36%), sensitivity (91.94%), specification (99.67%), MCC (0.895) and the best Kappa value (0.889) were obtained in the KNN-NCA prediction method. © 2024 IEEE.

关键词： Hepatitis C Disease KNN NCA Prediction SVM

来源：评论

学校读者我要写书评

暂无评论

Exploring Shifted Domain Problem within MNIST Dataset

Exploring Shifted Domain Problem within MNIST Dataset

引用

2023 International Conference on computer science, Information technology and Engineering, ICCoSITE 2023

作者： Hidayaturrahman Heryadi, Yaya Lukas Suparta, Wayan Arifin, Yulyani Bina Nusatara University BINUS Graduate Program - Computer Science Computer Science Department Jakarta11480 Indonesia Faculty of Engineering Jakarta12930 Indonesia Institut Teknologi Nasional Yogyakarta Faculty of Industrial Technology Department of Electrical Engineering Yogyakarta55281 Indonesia

ISBN: (纸本)9798350320954

Domain data can be shifted in any direction so it will be shared in different distributions to its original domain. This could be a problem since the model was trained with different distributions. It is found that adversarial domain adaptation using domain adversarial neural networks (DANN) can help to solve this problem on some scale. DANN can minimize the discrepancy between source and target data so the model can work well in both domains. The experiment is done by utilizing MNIST dataset that shifted into some conditions. In a condition when the shifting of distribution is too far, DANN is struggling to maintain the knowledge extracted from source data which leads to underperformance in the source and target domain. In contrast, when the shifting is closer, DANN can easily fit the model so it can perform well in both domains. It proves DANN is one of the good approaches to performing domain adaptation in small discrepancies. © 2023 IEEE.

关键词： Adversarial Domain Adaptation Domain Shifting Problem MNIST Transfer Learning

来源：评论

学校读者我要写书评

暂无评论

Production and characterization of polymeric nanocomposite foams based on starch and cellulose nanofibers from oil palm mesocarp fiber

引用

Applied Nanoscience (Switzerland) 2025年第2期15卷 1-11页

作者： Souza, Mariana Costa Gonçalves, Ana Paula Bispo da Silva Oliveira, Emanoel Igor dos Santos, Juscivaldo Passos Lazzari, Lídia Kunz Zattera, Ademir José José, Nadia Mamede Chemical Engineering Graduate Program Polytechnic School Federal University of Bahia Bahia Salvador Brazil Department of Education Federal Institute of Education Science and Technology Feira de Santana Bahia Brazil Postgraduate Program in Process Engineering and Technologies University of Caxias Do Sul Rio Grande Do Sul Caxias Do Sul Brazil

The demand for sustainable and eco-friendly materials has promoted studies over the years to explore different polymeric materials that meet requirements such as biodegradability and sustainability. In this context, biopolymer materials based on cellulose nanoparticles and starch from different botanical sources have been investigated, aiming to achieve satisfactory performance. The present study aims to develop and characterize cassava starch foams based on density, scanning electron microscopy (SEM), and three-point flexure tests, and to evaluate the effect of cellulose nanofibers obtained from palm mesocarp fibers using the ultrafine friction grinding method after different processing times, characterized by X-ray diffractometry (XRD), microscopy, and chemical composition, as a reinforcing filler. The chemical composition and scanning electron microscopy revealed the effectiveness of the cellulose isolation process, with the analysis of the chemical composition revealing a cellulose content of 55.70% in the fibers after pulping and bleaching, in addition to changes in the visual characteristics of the material after the process aimed at isolating the cellulose. After grinding, XRD showed an increase in crystallinity (76.1% at the maximum grinding time), along with typical microscopy images of cellulose nanofibers. The analysis of the obtained nanocomposites provided insights into the role of these nanostructures in the thermo-expanded starch matrix, indicating that the nanofibers promoted changes such as an increase in mechanical properties and crystallinity, which contributed to improving overall mechanical performance. A 67.48% increase in flexural strength was achieved for the formulation with cellulose nanofibers that underwent 150 min of grinding, without causing major variations in density. © King Abdulaziz City for science and technology 2025.

关键词： Starch

来源：评论

学校读者我要写书评

暂无评论

Benchmarking Performance of Document Level Classification and Topic Modeling

引用

computers, Materials & Continua 2022年第4期71卷 125-141页

作者： Muhammad Shahid Bhatti Azmat Ullah Rohaya Latip Abid Sohail Anum Riaz Rohail Hassan Department of Computer Science COMSATS University IslamabadLahorePakistan Department of Communication Technology and Network Faculty of Computer ScienceUniversiti Putra MalaysiaSelangor43400Malaysia Othman Yeop Abdullah Graduate School of Business(OYAGSB) Universiti Utara Malaysia(UUM)Kuala Lumpur50300Malaysia

Text classification of low resource language is always a trivial and challenging *** paper discusses the process of Urdu news classification and Urdu documents *** is one of the most famous spoken languages in *** implementation of computational methodologies for text classification has increased over ***,Urdu language has not much experimented with research,it does not have readily available datasets,which turn out to be the primary reason behind limited research and applying the latest methodologies to the *** overcome these obstacles,a mediumsized dataset having six categories is collected from authentic Pakistani news *** is a rich but complex *** processing can be challenging for Urdu due to its complex features as compared to other *** frequency-inverse document frequency(TFIDF)based term weighting scheme for extracting features,chi-2 for selecting essential features,and Linear discriminant analysis(LDA)for dimensionality reduction have been *** matrix and cosine similarity measure have been used to identify similar documents in a collection and find the semantic meaning of words in a document FastText model has been *** training-test split evaluation methodology is used for this experimentation,which includes 70%for training data and 30%for testing ***-of-the-art machine learning and deep dense neural network approaches for Urdu news classification have been ***,we trained Multinomial Naïve Bayes,XGBoost,Bagging,and Deep dense neural *** and deep dense neural network outperformed the other *** experimental results show that deep dense achieves 92.0%mean f1 score,and Bagging 95.0%f1 score.

关键词： Deep neural network machine learning natural language processing TFIDF sparse matrix cosine similarity classification linear discriminant analysis gradient boosting

来源：评论

学校读者我要写书评

暂无评论

Framework for Integration of Generative AI into Metaverse Asset Creation

Framework for Integration of Generative AI into Metaverse As...

引用

Intelligent Metaverse Technologies & Applications (iMETA), International Conference on

作者： Rio Mukhtarom Paweroi Mario Köppen Department of Computer Science and Networks Graduate School of Computer Science and Systems Engineering Kyushu Institute of Technology Iizuka Japan

ISBN: (数字)9798350351514

ISBN: (纸本)9798350351521

Metaverse, a virtual world, is developing rapidly and is widely used in multi-sector. The number of users is projected to increase year over year. Due to the development of the metaverse platform, digital asset creation is demanding. Creating high-quality and diverse 3D digital objects is challenging. This study proposes the frameworks for integrating generative AI to create diverse 3D assets into the metaverse. We study different approaches for asset creation, i.e., generative 3D model-based, generative image projection-based, and generative language script-based. Creators can use this workflow to optimize the creation of 3D assets. Moreover, this study compares the results of generative AI and procedural generation on generating diverse 3D objects. The result shows that generative AI can simplify 3D creation and generate more diverse objects.

关键词： Solid modeling Three-dimensional displays Procedural generation Generative AI Metaverse Complexity theory Usability

来源：评论

学校读者我要写书评

暂无评论

Examining the Impact of Expression Fragmentation and Number Redundancy on Financial QA task

Examining the Impact of Expression Fragmentation and Number ...

引用

2025 IEEE International Conference on Big Data and Smart Computing, BigComp 2025

作者： Lee, Yoonjae Ki, Kyung Seo Gweon, Gahgene Kim, Bugeun Seoul National University Graduate School of Convergence Science and Technology Korea Republic of Seoul National University Interdisciplinary Program in Artificial Intelligence Korea Republic of Chung-Ang University Department of AI Korea Republic of

The goal of financial QA is to generate solution equations by solving problems about financial reports. Current financial QA models can suffer from two issues: expression fragmentation and number redundancy. We conduct experiments to examine the impact of the two issues. The experimental results show that addressing the expression fragmentation issue in financial QA using EPT improves the execution accuracy by 0.09, and alleviating the number redundancy issue by removing redundant numbers in the input of the FinQANet generator improves the execution accuracy by 0.07. © 2025 IEEE.

关键词： Carrier sense multiple access

来源：评论

学校读者我要写书评

暂无评论

Formal Security Verification for Searchable Symmetric Encryption Using ProVerif

Formal Security Verification for Searchable Symmetric Encryp...

引用

International Symposium on Information Theory and Its Applications, ISITA

作者： Takehiko Mieno Hiroyuki Okazaki Kenichi Arai Yuichi Futa Hiroaki Yamamoto Graduate School of Medicine Science and Technology Shinshu University EPSON AVASYS CORPORATION Graduate School of Science and Technology Shinshu University Graduate School of Engineering Nagasaki University School of Computer Science Tokyo University of Technology Faculty of Engineering Shinshu University

ISBN: (数字)9784885523519

ISBN: (纸本)9798331533335

With the rapid proliferation of various cloud storage services in recent years, the development of technology to efficiently search data while ensuring its confidentiality during cloud usage is an important issue. The technology that enables keyword searches on encrypted files using previously set keywords is called searchable symmetric encryption (SSE). In this paper, we propose a method formally representing encrypted document, and verify the security of SSE using the formal verification tool ProVerif. Our proposed method considers the channel-type terms of ProVerif as a Document that includes different keywords to verify the indistinguishability of encrypted documents.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Smartphone Application for the Deaf and the Deaf Caring Community 6

Smartphone Application for the Deaf and the Deaf Caring Comm...

引用

6th International Conference on Inventive Computation Technologies, ICICT 2023

作者： Pryandi, Pinnus Dewantara, Muhammad Bayu Warnars, Harco Leslie Hendric Spits Arief, Ramadhan Information Systems Department School of Information Systems Bina Nusantara University Jakarta11480 Indonesia Computer Science Department BINUS Graduate Program Doctor of Computer Science Bina Nusantara University Jakarta11480 Indonesia Telkom University Bandung11480 Indonesia

ISBN: (纸本)9798350398496

In life, various challenges and problems faced by deaf people such as communication skills and other problems, including emotional, mental, and societal development. However, technology is needed that can help the process. This Smart Application try to build communication between the deaf to share thoughts in group chats, including getting information about the deaf, so that the communication learning process for the deaf becomes easier. The proposed model is using Unified Modeling Languages (UML) diagrams, such as case diagrams for the current proposed idea process and class diagrams for relational tables or database stores. In addition, the User interface (UI) is implemented in this research work, Personal Home Pages (PHP) as a web server programming and MySQL database as an open-source database. © 2023 IEEE.

关键词： User interfaces

来源：评论

学校读者我要写书评

暂无评论

Model of Text-to-Text Transfer Transformer for Hadith Question Answering System

Model of Text-to-Text Transfer Transformer for Hadith Questi...

引用

International Conference on Information technology, computer and Electrical Engineering (ICITACEE)

作者： Bambang Sulistio Derwin Suhartono Haryono Soeparno Widodo Budiharto Computer Science Department BINUS Graduate Program - Doctor of Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Doctor of Computer Science Program Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798350389302

ISBN: (纸本)9798350389319

The advancement of Artificial Intelligence (AI), notably in Natural Language Processing (NLP), has been remarkable. Among its applications, Question-Answering (QA) systems stand out, assisting users in accessing pertinent information across various topics, including religious contexts. Religion plays a significant role in providing guidance on moral values to modern society, such as tolerance, compassion and social norms. Religious practices taught by religion serve as significant foundations for addressing the increasingly advancing influence of technology. The primary goal of this paper is to create a question-answering system proficient in analyzing user inquiries and precisely extracting data from translated Indonesian Hadith datasets. This research endeavors to enhance question-answering accuracy through Deep Learning techniques. Employing pre-processing methods, the system interprets user query intents. Furthermore, the Text-to-Text Transfer Transformer (T5), a text-based language Transformer model, is utilized to streamline the retrieval process for Hadith-related queries based on relevant subjects. The research findings demonstrate the precision of answers through BLEU and ROUGE scores. The novelty of this study lies in the creation of a dataset of hadith translations in Indonesian and the fine-tuning of the T5 model that has been trained using a dataset of questions and answers based on these translations. This research demonstrates the following evaluation scores: BLEU: 0.606, ROUGE-1: 0.702, ROUGE-2: 0.548, and ROUGE-L: 0.701, indicating that the answers generated by this particular research model are better compared to those produced using the standard mT5 model.

关键词： Electrical engineering Deep learning Ethics Large language models Transformers Question answering (information retrieval) Data mining Information technology Standards Testing

来源：评论

学校读者我要写书评

暂无评论

Unifying Emotion-Oriented and Cause-Oriented Predictions for Emotion-Cause Pair Extraction

SSRN

引用

SSRN 2024年

作者： Hu, Guimin Zhao, Yi Lu, Guangming School of Computer Science and Technology Harbin Institute of Technology Shenzhen Graduate School Shenzhen518000 China School of Science Harbin Institute of Technology Shenzhen Graduate School Shenzhen518000 China

Emotion-cause pair extraction (ECPE) is an extraction task aiming to simultaneously identify the emotions and causes from the text without emotion annotations. Let $c_{i}$ and $c_{j}$ be the emotion clause and the cause clause of a document, respectively, and set we can predict one from the other and vice versa. Previous works fail to take advantage of this bidirectional opportunity. We refer to the prediction from $c_{i}$ to $c_{j}$, i.e., $c_{i}\rightarrow c_{j}$, as an emotion-oriented cause prediction (EoCP) task and the prediction from $c_{j}$ to $c_{i}$, i.e., $c_{j}\rightarrow c_{i}$, as a cause-oriented emotion prediction (CoEP) task. After the new decomposition of the ECPE task, we propose a novel unified architecture for ECPE, which takes EoCP and CoEP as cells and unifies them into a single-chain architecture. Meanwhile, we redefine emotion-cause pair extraction as a closed-loop structure detection problem to alleviate the mismatch between emotion and cause clauses. To better train the architecture, we define a new loss and provide a procedure for estimating the confidence the extraction system has for its emotion-cause pairs. We demonstrate the superiority of our proposed model through extensive experiments on two public datasets and reap a new state-of-the-art performance. Our method especially obtains significant improvement in multiple emotion-cause pair extraction. © 2024, The Authors. All rights reserved.

关键词： Extraction

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：