检索结果-内蒙古大学图书馆

Performance Evaluation of Coffee Bean Binary Classification Through Deep Learning Techniques 1

International Conference on Innovative Technologies in Intelligent Systems and Industrial Applications, CITISIA 2022

作者： Islamy, Fajrul Muchtar, Kahlil Arnia, Fitri Dawood, Rahmad Febriana, Alifya Elwirehardja, Gregorius Natanael Pardamean, Bens Department of Electrical and Computer Engineering Universitas Syiah Kuala Banda Aceh23111 Indonesia Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia Computer Science Department Master of Computer Science Program BINUS Graduate Program Bina Nusantara University Jakarta11480 Indonesia

ISBN: (数字)9783031290787

ISBN: (纸本)9783031290770

Coffee beans are one of the high-value commodities in Indonesia, but the sorting method for the quality of coffee beans still uses visual methods and sieves with mechanical machines. This study aims to provide an alternative to classifying coffee beans using Deep Learning algorithms, such as the ResNet-18 and MobileNetV2. The dataset used is USK-COFFEE which consists of two classes, the normal coffee bean class, and the defect coffee bean class. The models were trained using the balanced and imbalanced datasets, meaning that two scenarios were performed. From the results of this study, the accuracy value for the ResNet-18 model was 91.50% when trained on the balanced dataset and 90.87% on the imbalanced dataset. It outperformed the MobileNetV2 architecture, which only achieved 0.8125 and 0.8112 accuracy scores when trained on the balanced and imbalanced datasets, respectively. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

The planning strategy of robotics technology for nuclear decommissioning in Taiwan

引用

Nuclear Engineering and Technology 2024年第1期56卷 64-69页

作者： Tu, Chung Yi Chen, Kuen Tsann Ting, Kuen Yang Sheng, Chin Doctoral Program in Data Science and Industrial Analytics National Chung Hsing University Taichung County 402 Taiwan Department of Applied Mathematics National Chung Hsing University Taichung County 402 Taiwan Department of Mechanical Engineering Lunghwa Univesity of Science and Technology Taoyuan County 33306 Taiwan

According to the market research report, the nuclear decommissioning services market is currently experiencing considerable growth, with a projected Compound Annual Growth Rate (CAGR) of nearly 13% during the 2020–2024 forecast period. This expansion is primarily fueled by the advancement of Industry 4.0, in conjunction with the emergence of cutting-edge technologies such as the Internet of Things, big data, artificial intelligence, and 5G. Even though the fact that robots have already been utilized in the nuclear industry, their adoption has been hindered by conservative regulations. However, the nuclear decommissioning market presents an opportunity for the advancement of robotics technology. The British have already invested heavily in encouraging the use of intelligent robots for nuclear decommissioning, and other countries, such as Taiwan, should follow suit. Taiwan's flourishing robotics development industry in manufacturing, logistics, and other domains can be leveraged to introduce advanced robotics in the decommissioning of its nuclear power plants. By doing so, Taiwan can establish itself as a competitive player in the nuclear decommissioning services market for the next two decades. © 2023 Korean Nuclear Society

关键词： Intelligent robotics Nuclear decommissioning Planning strategy

来源：评论

学校读者我要写书评

暂无评论

Implications of the Usage of Three-Mode Principal Component Analysis with a Fixed Polynomial Basis 15th

Implications of the Usage of Three-Mode Principal Compone...

引用

15th International KES Conference on Intelligent Decision Technologies, KESIDT 2023

作者： Monden, Rei Nagai, Isamu Yanagihara, Hirokazu Informatics and Data Science Program Graduate School of Advanced Science and Engineering Hiroshima University Hiroshima Higashi-Hiroshima739-8521 Japan Faculty of Liberal Arts and Science Chukyo University Aichi Nagoya466-8666 Japan Mathematics Program Graduate School of Advanced Science and Engineering Hiroshima University Hiroshima Higashi-Hiroshima739-8526 Japan

ISBN: (纸本)9789819929689

When I subjects answer questions regarding J variables K times, the data can be stored in a three-mode data set of size I× J× K. Among the various component analysis approaches to summarize such data, Three-mode Principal Component Analysis (3MPCA) is often used. 3MPCA expresses a three-mode data set by means of A-, B-, and C-mode components and a core array, but this approach is not suitable for predicting data at unobserved time points. To handle this issue, the current study suggests a fixed orthonormal polynomial basis for the 3MPCA model. Briefly, we fix the C-mode component matrix with an orthonormal polynomial basis. More specifically, i) measurement time points, and ii) log transformation of the measurement time points were used as an orthonormal polynomial basis. The use of our proposed methods was demonstrated by applying them to a longitudinal data set collected from patients with depression. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Estimation Algorithms for MLE of Three-Mode GMANOVA Model with Kronecker Product Covariance Matrix 15th

Estimation Algorithms for MLE of Three-Mode GMANOVA Model ...

引用

15th International KES Conference on Intelligent Decision Technologies, KESIDT 2023

作者： Horikawa, Keito Nagai, Isamu Monden, Rei Yanagihara, Hirokazu Informatics and Data Science Program Graduate School of Advanced Science and Engineering Hiroshima University Hiroshima Higashi-Hiroshima739-8521 Japan Faculty of Liberal Arts and Science Chukyo University Aichi Nagoya466-8666 Japan Mathematics Program Graduate School of Advanced Science and Engineering Hiroshima University Hiroshima Higashi-Hiroshima739-8526 Japan

ISBN: (纸本)9789819929689

Currently, there are a lot of measurement data on different items collected over time. The GMANOVA model is appropriate for analyzing the trends in such data, in order to analyze some longitudinal data collected on different items simultaneously, we need to create separate GMANOVA models, but separate GMANOVA models cannot reveal the relationships between items. Also, if we try to apply one model to all of these data, the problem arises that there are too many parameters and the covariance structure is too large. Therefore, we propose an extended GMANOVA model and two estimation algorithms. The proposed model’s covariance matrix has a Kronecker structure. Therefore, because the estimator cannot be uniquely determined using the log-likelihood, we propose two parameter estimation algorithms. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Covariance matrix

来源：评论

学校读者我要写书评

暂无评论

Fetal Health Risk Classification Using Important Feature Selection and Cart Model on Cardiotocography data

Informatica (Slovenia)

引用

Informatica (Slovenia) 2025年第1期49卷 193-206页

作者： Ilham, Ahmad Nagara, Thahta Ardhika Prabu Kamaruddin, Mudyawati Khikmah, Laelatul Mantoro, Teddy Department of Informatics Universitas Muhammadiyah Semarang Semarang Indonesia Intelligent Data Science Research Group Universitas Muhammadiyah Semarang Semarang Indonesia Postgraduate Program of Medical Laboratory Science Universitas Muhammadiyah Semarang Semarang Indonesia Interdisciplinary Research Laboratory for Experimental Plasma Medicine Universitas Muhammadiyah Semarang Semarang Indonesia Department of Statistics Institut Teknologi Statistika dan Bisnis Muhammadiyah Semarang Semarang Indonesia School of Computer Science Nusa Putra University West Java Indonesia

Fetal mortality and newborn health issues require urgent attention because of high maternal and infant mortality rates during labor, highlighting the critical need for accurate fetal condition monitoring to reduce complications. This study proposes the development of a fetal health risk classification model based on Important Feature Selection (IFS) and a Classification and Regression Tree (CART) using cardiotocography (CTG) data from the UCI Machine Learning Repository. The IFS method was used to select the most relevant features, reduce model complexity, and increase generalization to prevent overfitting. The IFS-CART model was tested with 10-fold cross-validation and showed an accuracy of 94.50%, superior to the conventional CART, which only reached 93.83%. In addition, the average values of the True Positive Rate (TPR) and True Negative Rate (TNR) also increased, indicating that this model is effective in distinguishing normal, suspected, and pathological fetal conditions. Evaluation using the area under the curve receiver operating characteristic (AUC-ROC) showed that the model had high performance in detecting at-risk conditions, with an AUC of 0.981 for the"suspect" class. This finding confirmed that IFS-CART is not only accurate but also has high interpretability, making it easy for medical personnel to use for clinical decision support. The results of this study show that IFS-CART can serve as a reliable decision support system for real-time fetal health monitoring. Further implementation is expected to improve diagnostic accuracy and prevent complications during pregnancy and labor © 2025 Slovene Society Informatika. All rights reserved.

关键词： Fetal monitoring

来源：评论

学校读者我要写书评

暂无评论

Phishing Email Detection Model Using Deep Learning

Phishing Email Detection Model Using Deep Learning

引用

2025 IEEE International Conference on Cybernetics and Innovations, ICCI 2025

作者： Pimpason, Natthaphong Viboonsang, Pranodnard Kosolsombat, Somkiat College of Interdisciplinary Studies Thammasat University Data Science and Innovation Program Pathum Thani Thailand

ISBN: (纸本)9798331533267

Phishing via email continues to be a concern in the cybersecurity realm as it exposes both individuals and companies to levels of risk due to clever social engineering methods employed by cybercriminals. This study introduces an extensive deep learning strategy for identifying phishing emails using a blend of various models such as Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional Encoder Representations from Transformers (BERT), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN). The suggested approach utilizes natural language processing and learning techniques to evaluate and categorize email material by identifying subtle attributes that differentiate valid from harmful communication effectively. Through testing and careful evaluation of the model performance, achieved results;the LSTM model attained a notable accuracy of 99.92% in detecting phishing emails. This progress highlights the promise of learning methods in addressing complex phishing threats and improving email security measures. The research offers insights into the effectiveness of neural network approaches in detecting phishing emails by comparing the performance of multiple deep learning architectures. The high level of accuracy attained indicates a path for creating smarter email security solutions that can evolve to counter cyber threats effectively. © 2025 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Digital Twin (DT) Smart City for Air Quality Management 8

Digital Twin (DT) Smart City for Air Quality Management

引用

8th International Conference on Computer science and Computational Intelligence, ICCSCI 2023

作者： Ariansyah, Dedy Isnan, Mahmud Rahutomo, Reza Pardamean, Bens Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia Information Systems Department School of Information Systems Bina Nusantara University Jakarta11480 Indonesia Computer Science Department BINUS Graduate Program-Master of Computer Science Program Bina Nusantara University Jakarta11480 Indonesia

Air pollution is a pressing issue in cities, and managing air quality poses a challenge for urban designers and decision-makers. This study proposes a Digital Twin (DT) Smart City integrated with Mixed Reality technology to enhance visualization and collaboration for addressing urban air pollution. The research adopts an applied research approach, with a focus on developing a DT framework. A use case of DT development for Jakarta, the capital of Indonesia, is presented. By integrating air quality data, meteorological information, traffic patterns, and urban infrastructure data, the DT provides a comprehensive understanding of air pollution dynamics. The visualization capabilities of the DT, utilizing Mixed Reality technology, facilitate effective decision-making and the identification of strategies for managing air quality. However, further research is needed to address data management challenges to build a DT for Smart City at scale. © 2023 The Authors. Published by Elsevier B.V.

关键词： Augmented Reality Digital Twin Mixed Reality Smart City

来源：评论

学校读者我要写书评

暂无评论

Time-series Analysis of Correlation between Climatic Parameters and Dengue Fever in Indonesia 4

Time-series Analysis of Correlation between Climatic Paramet...

引用

4th International Conference on Informatics, Multimedia, Cyber and Information System, ICIMCIS 2022

作者： Nirwantono, Rudi Trinugroho, Joko Pebrianto Sudigyo, Digdo Hidayat, Alam Ahmad Pardamean, Bens Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia Bina Nusantara University Binus Graduate Program Master of Computer Science Program Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9781665473279

Dengue fever is mainly caused by female mosquitoes from the genus Aedes. It predominantly occurs in equatorial to sub-tropical regions. The weather factors play a significant role in the mosquito life cycle, so a study on the impact of climate parameters on dengue incidence is required. This study aimed to understand the statistical association between the annual fluctuation of dengue cases with climate factors in Indonesia. Climatic data, annual rainfall, and annual mean ambient temperature of Indonesia were obtained from *** from 2010 to 2020, while the data on the annual incidence of dengue fever in Indonesia were collected from the Indonesian Health Profile by the Ministry of Health of Indonesia from 2010 to 2020. The association between climatic factors with dengue cases was calculated with Spearman statistical test after the data was indicated as not normally distributed. In addition, we performed Poisson regression analysis to examine how the changes in climatic parameters affect the dengue cases. The rain precipitation was insignificantly correlated with the dengue incidence in Indonesia during the study time. In contrast, the ambient temperature statistically showed a strong association with the dengue cases. The association between climatic factors and the number of dengue cases suggests that the model would likely be prospectively applied to the dengue early warning system as a nationwide intervention to reduce the epidemic impact in Indonesia. © 2022 IEEE.

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

Phishing Email Detection Model Using Deep Learning

Phishing Email Detection Model Using Deep Learning

引用

Cybernetics and Innovations (ICCI), International Conference on

作者： Natthaphong Pimpason Pranodnard Viboonsang Somkiat Kosolsombat Data Science and Innovation Program College of Interdisciplinary Studies Thammasat University Pathum Thani Thailand

ISBN: (数字)9798331533267

ISBN: (纸本)9798331533274

Phishing via email continues to be a concern in the cybersecurity realm as it exposes both individuals and companies to levels of risk due to clever social engineering methods employed by cybercriminals. This study introduces an extensive deep learning strategy for identifying phishing emails using a blend of various models such as Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional Encoder Representations from Transformers (BERT), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN). The suggested approach utilizes natural language processing and learning techniques to evaluate and categorize email material by identifying subtle attributes that differentiate valid from harmful communication effectively. Through testing and careful evaluation of the model performance, achieved results; the LSTM model attained a notable accuracy of 99.92% in detecting phishing emails. This progress highlights the promise of learning methods in addressing complex phishing threats and improving email security measures. The research offers insights into the effectiveness of neural network approaches in detecting phishing emails by comparing the performance of multiple deep learning architectures. The high level of accuracy attained indicates a path for creating smarter email security solutions that can evolve to counter cyber threats effectively.

关键词： Deep learning Accuracy Phishing Feature extraction Natural language processing Encoding Electronic mail Security Convolutional neural networks Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Optimizing Malware Detection with Random Forest, XGBoost, LightGBM, and LLM-Reporting

Optimizing Malware Detection with Random Forest, XGBoost, Li...

引用

Cybernetics and Innovations (ICCI), International Conference on

作者： Aticha Charoenthanakitkul Pranodnard Viboonsang Somkiat Kosolsombat Data Science and Innovation Program College of Interdisciplinary Studies Thammasat University Pathum Thani Thailand

ISBN: (数字)9798331533267

ISBN: (纸本)9798331533274

This study presents a comprehensive framework for malware detection that integrates traditional machine learning algorithms with advanced large language models (LLMs) to enhance both classification accuracy and automated report generation. The proposed approach leverages three widely used machine learning models-Random Forest, XGBoost, and LightGBM-to classify different types of malwares based on extracted features from the EMBER dataset, a well-known benchmark dataset for malware analysis. Each of these models was trained and evaluated to determine their effectiveness in identifying malicious software with high accuracy. Among the tested algorithms, LightGBM exhibited the best performance, achieving an impressive classification accuracy of 83%. XGBoost followed closely with 81%, while Random Forest achieved an accuracy of 79%. To further enhance the usability of the system for cybersecurity analysts, the study incorporates a large language model, specifically LLaMA, which is accessed via an API. This model is employed to generate detailed, human-readable reports that provide insights into detected malware threats. By automating the reporting process, the framework aims to reduce the manual effort required for malware analysis, allowing security professionals to focus on high-priority threats. Future research will focus on improving classification performance through advanced feature engineering techniques and exploring deep learning-based neural network architectures to further enhance the accuracy and efficiency of malware detection and analysis.

关键词： Training Analytical models Accuracy Machine learning algorithms Computational modeling Large language models Feature extraction Malware Computer security Random forests

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：