检索结果-内蒙古大学图书馆

6th International Conference on Cybernetics and Intelligent System, ICORIS 2024

作者： Sari, Azani Cempaka Heryadi, Yaya Kartowisastro, Iman Herwidiana Budiharto, Widodo Computer Science Department BINUS Graduate Program Bina Nusantara University Jakarta11480 Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia

ISBN: (纸本)9798331539603

This paper investigates recent advancements and methodologies in Automatic Text Summarization, a field that has garnered significant attention for its ability to condense lengthy documents into concise summaries while preserving key information and themes. The study examines the development of Automatic Text Summarization models, including both extractive and abstractive approaches. Additionally, the research delves into the use of graph convolutional networks for text representation in e-commerce product review summarization. The study employs an e-commerce dataset in conjunction with a graph convolutional network model, covering stages from preprocessing and tokenization to constructing graph-based data representation. The expected outcome is a graph-structured representation of input text data derived from Indonesian-language e-commerce product reviews. © 2024 IEEE.

关键词： Convolutional E-commerce Graph Network Product Review Summarization Text

来源：评论

学校读者我要写书评

暂无评论

Comparative Study to Find the Best Deep Learning Feature Extraction Models and Machine Learning Classification Algorithm to Detect Monkeypox Disease

Comparative Study to Find the Best Deep Learning Feature Ext...

引用

2025 International Conference on Advancement in data science, E-learning and Information System, ICADEIS 2025

作者： Abdullah, Hasna Salsabilla Tjitra, Tammy Visca Kwandinata, Vanessa Achmad, Said School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798331513320

Monkeypox disease that has spread throughout the world for the past few years has a form similar to other diseases, making it difficult to identify monkeypox correctly and patients end up getting the wrong or less effective treatment. Deep Learning is the right solution to detect monkeypox because of its ability to handle complex and varied data, such as medical images. With CNN architecture, Deep Learning can automatically learn deep features directly from data without manual extraction, making it effective in recognizing complex patterns that are difficult for humans to identify. This study aims to find the best Deep Learning and classification model to distinguish monkeypox from other similar diseases (acne, chickenpox, cowpox, hand foot and mouth (HFMD), and measles). The models tested are CNN models: VGG16Net, MobileNetV2, and DenseNet-201 which have been fine-tuned combined with the Random Forest classification model with input in the form of human skin images. Each model is evaluated by calculating the confusion matrix (accuracy, precision, recall, and F-1 Score) and AUC ROC to determine the best performance. The highest accuracy was produced by the MobileNetV2 and SVM models, reaching 94.41%, while the highest AUC ROC average result was produced by the DenseNet- 201 and SVM models at 98.67%. This study shows that Deep Learning models have great potential to help diagnose diseases accurately. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Teacher Probability Reconstruction based knowledge distillation within intelligent network compression

引用

International Journal of Intelligent Networks 2025年 6卷 47-56页

作者： Chen, Han Teng, Xuyang Su, Jiajie Li, Chunhao Hu, Chang Han, Meng School of Communication Engineering Hangzhou Dianzi University Hangzhou China Department of Computer Science National University of Singapore Singapore College of Computer Science and Technology Zhejiang University Hangzhou China

In the optimization of intelligent network architecture, limited resources at each node, including edge computing devices, have posed challenges for deploying large models in performance-demanding scenarios. Knowledge distillation serves as a model compression method that extracts knowledge from a large-scale teacher model and transfers it to a more lightweight student model. Previous knowledge distillation methods mainly focus on the intermediate layers of the network. However, due to privacy protection regulations that limit data sharing and access as well as computational efficiency requirements in practical scenarios, feature-based distillation encounters challenges in practical applications. We start with logit-based distillation to address these issues, enabling students to learn more representative knowledge from the teacher's output probability distribution. Due to the structural limitations of the teacher network such as insufficient depth or width, and potential issues in the training data like noise and imbalance, the output probability distribution contains many errors. Therefore, we propose a knowledge distillation method that improves the student by correcting errors in the teacher model. Nevertheless, teacher's errors not only bring mistakes to students but also give students greater subjectivity, enabling them to break free from the limitations of the teacher. We also retain the teacher's thinking to prevent students from becoming biased on the remaining (non-target) categories while correcting teacher errors for students. Extensive experiments demonstrate that our method achieves competitive performance on multiple benchmarks without extra parameters. © 2025 The Authors

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Ensemble Techniques for Bone Fracture Detection from X-Ray Images Using HOG Features

Comparative Analysis of Ensemble Techniques for Bone Fractur...

引用

2025 International Conference on Advancement in data science, E-learning and Information System, ICADEIS 2025

作者： Muzakki, Abyan Farrel Prasetyo, Simeon Yuda Purwanto, Eko Setyo School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798331513320

Musculoskeletal conditions encompass 150 varying conditions and approximately affect 1.71 billion people worldwide, according to the World Health Organization. Bone fractures are among the musculoskeletal conditions that significantly impact individuals, families, societies, and healthcare systems, as they can lead to work absences, decreased productivity, disability, and a host of other life-limiting conditions. This study aims to evaluate the performance of superior-performing machine learning (ML) models that are fed with histograms of oriented gradients (HOG). The bestperforming ML models were chosen and implemented toward ensemble models in aims to reduce overfitting and address resource-limited settings. The results indicate that by using KNearest Neighbors (KNN), Decision Tree (DT), and Random Forest (RF), which yield the highest scores, the proposed ensemble models were able to increase the score, with the stacking ensemble model achieving accuracy score of 99.30%. Future research could focus on integrating advanced machine learning techniques, hybrid feature extraction methods, and larger datasets to enhance diagnostic accuracy and applicability in clinical environments. © 2025 IEEE.

关键词： Musculoskeletal system

来源：评论

学校读者我要写书评

暂无评论

Combining DistilBERT and LSTM to Enhanced Detection of Depression in Social Media Text 9

Combining DistilBERT and LSTM to Enhanced Detection of Depre...

引用

9th International Conference on Information Management and Technology, ICIMTech 2024

作者： Kurniadi, Felix Indra Satyaning Pradnya Paramita, Ni Luh Putu Angela Sihotang, Erna Fransisca Anggreainy, Maria Susan Zhan, Rongfang School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia School of Computer Science Bina Nusantara University Statistics Department Jakarta11480 Indonesia BINUS Graduate Program - Doctor of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia School of Computer Science University of North Texas Department of Rehabilitation and Health Service United States

ISBN: (纸本)9798350390025

Depression affects a significant number of people, and it has a significant impact not only on their lives but also on society as a whole. In light of this, we require more advanced methods that are capable of locating individuals who are depressed in a timely and accurate manner. This research investigates the effectiveness of Long Short-Term Mem- ory (LSTM) networks used in conjunction with DistilBERT, a small transformer-based model, in identifying indications of melancholy in posts made on social networking platforms. The text was cleaned up using a dataset that was obtained from Reddit, and then we used the mixed DistilBERT+LSTM model to evaluate how well it performed in comparison to the DistilBERT model by itself. The findings that we obtained indicate that both approaches produce results that are comparable but not flawless, with only slight variations in terms of accuracy, precision, recall, and F1 scores. The fact that this is the case demonstrates that there are issues with optimizing the model and displaying the data. © 2024 IEEE.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Comparison of Spelling Error Correction Algorithms for the Indonesian Language 11

Comparison of Spelling Error Correction Algorithms for the I...

引用

11th International Conference on Information and Education Technology, ICIET 2023

作者： Yanfi, Yanfi Setiawan, Reina Soeparno, Haryono Budiharto, Widodo Bina Nusantara University BINUS Graduate Program Computer Science Jakarta Indonesia Bina Nusantara University BINUS Graduate Program Computer Science School of Computer Science Computer Science Department Jakarta Indonesia Bina Nusantara University School of Computer Science Computer Science Department Jakarta Indonesia

ISBN: (纸本)9781665465489

Edit distance as a string measurement metric is often used to help detect misspellings in languages. This paper aims to compare two string spelling error correction algorithms for the Indonesian language. The N-gram, Jaro-Winkler distance, and Levenshtein distance algorithms are used to determine whether they can accurately correct typological errors in the Indonesian language. Moreover, this study utilized KNIME tools to process the data from beginning to end. The data was retrieved from news in Indonesia. After the experiment on N from 1 to 12, the results obtained for the comparative analysis proved that Jaro-Winkler distance performed better than Levenshtein distance for comparing smaller strings like words and names. However, Levenshtein distance performs as well as Jaro-Winkler distance started from four strings. Finally, both Jaro-Winkler distance and Levenshtein distance algorithm got the best performance accuracy for eight strings with an accuracy of 99.52 percent. The result of this study is also presented that both algorithms can support word error correction for the Indonesian language. © 2023 IEEE.

关键词： Error correction

来源：评论

学校读者我要写书评

暂无评论

Fast and scalable off-target assessment for CRISPR guide RNAs using partial matches

Fast and scalable off-target assessment for CRISPR guide RNA...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Schmitz, Carl Bradford, Jacob Salomone, Robert Perrin, Dimitri Queensland University of Technology School of Computer Science Centre for Data Science Brisbane Australia

ISBN: (纸本)9798350386226

The design of CRISPR-Cas9 guide RNAs requires the analysis of all potential target sites to determine the risk of unintentional edits for any given guide. A number of methods have been developed to evaluate this risk but the size of the target genome is a challenging factor. Many existing methods are limited by their need for an unfeasible amount of time or memory. In this paper, we introduce Crackling++, an update to our previous Crackling method, in which we have reduced processing time by imposing stricter matching criteria. We conducted a benchmark to compare our performance to other tools and show that Crackling++ is the fastest off-targeting scoring method and that it can scale to the largest genomes. © 2024 IEEE.

关键词： Genome

来源：评论

学校读者我要写书评

暂无评论

Enhancing Emotion Detection With Bidirectional GRU Networks: A Deep Learning Approach Using Dropout and Batch Normalization

Enhancing Emotion Detection With Bidirectional GRU Networks:...

引用

2025 International Conference on Advancement in data science, E-learning and Information System, ICADEIS 2025

作者： Luckianto, Marvin Addison, Renaldi Santoso, Alexander Agung Gunawan School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798331513320

Social media platforms, such as Twitter, produce significant quantities of textual data, providing important opportunities for the analysis of emotions. The identification of emotions within text is essential for applications including mental health monitoring, marketing strategies, and sentimentdriven recommendation systems. Traditional methods frequently encounter challenges associated with linguistic complexities and the ability to generalize across varied datasets. This paper proposes a novel method using Bidirectional Gated Recurrent Unit (BiGRU) networks with some regularization techniques such as dropout and batch normalization. The bidirectional architecture allows both forward and backward contextual information to be learned, thus it can express subtle emotional features effectively. Using a dataset of 89,832 labeled tweets from six different emotional classes - sadness, joy, love, anger, fear, and surprise - the model achieved an accuracy of 93.83%, precision of 93.83%, recall of 93.83%, and an F1-score of 93.83%. This demonstrates that integrating advanced neural architectures along with successful regularization techniques significantly improves the stability and effectiveness of emotion detection models. © 2025 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Multi-head attention graph convolutional network model:End-to-end entity and relation joint extraction based on multi-head attention graph convolutional network

引用

CAAI Transactions on Intelligence Technology 2023年第2期8卷 468-477页

作者： Zhihua Tao Chunping Ouyang Yongbin Liu Tonglee Chung Yixin Cao School of Computer Science University of South ChinaHengyangChina Department of Computer Science and technology Tsinghua UniversityBeijingChina School of Computer and Information Systems Singapore Management UniversitySingaporeSingapore

At present,the entity and relation joint extraction task has attracted more and more scholars'attention in the field of natural language processing(NLP).However,most of their methods rely on NLP tools to construct dependency trees to obtain sentence structure *** adjacency matrix constructed by the dependency tree can convey syntactic *** trees obtained through NLP tools are too dependent on the tools and may not be very accurate in contextual semantic *** the same time,a large amount of irrelevant information will cause *** paper presents a novel end-to-end entity and relation joint extraction based on the multihead attention graph convolutional network model(MAGCN),which does not rely on external *** generates an adjacency matrix through a multi-head attention mechanism to form an attention graph convolutional network model,uses head selection to identify multiple relations,and effectively improve the prediction result of overlapping *** authors extensively experiment and prove the method's effectiveness on three public datasets:NYT,WebNLG,and *** results show that the authors’method outperforms the state-of-the-art research results for the task of entities and relation extraction.

关键词： information retrieval natural language processing

来源：评论

学校读者我要写书评

暂无评论

Elevating Image Steganography:A Fusion of MSB Matching and LSB Substitution for Enhanced Concealment Capabilities

引用

computers, Materials & Continua 2024年第5期79卷 2923-2943页

作者： Muhammad Zaman Ali Omer Riaz Hafiz Muhammad Hasnain Waqas Sharif Tenvir Ali Gyu Sang Choi Department of Information Technology The Islamia University of BahawalpurBahawalpur63100Pakistan Department of Computer Science The Islamia University of BahawalpurBahawalpur63100Pakistan School of Computer Science and Engineering Yeungnam UniversityGyeongbuk38541Korea

In today’s rapidly evolving landscape of communication technologies,ensuring the secure delivery of sensitive data has become an essential *** overcome these difficulties,different steganography and data encryption methods have been proposed by researchers to secure *** of the proposed steganography techniques achieve higher embedding capacities without compromising visual imperceptibility using LSB *** this work,we have an approach that utilizes a combinationofMost SignificantBit(MSB)matching andLeast Significant Bit(LSB)*** proposed algorithm divides confidential messages into pairs of bits and connects them with the MSBs of individual pixels using pair matching,enabling the storage of 6 bits in one pixel by modifying a maximum of three *** proposed technique is evaluated using embedding capacity and Peak Signal-to-Noise Ratio(PSNR)score,we compared our work with the Zakariya scheme the results showed a significant increase in data concealment *** achieved results of ourwork showthat our algorithmdemonstrates an improvement in hiding capacity from11%to 22%for different data samples while maintaining a minimumPeak Signal-to-Noise Ratio(PSNR)of 37 *** findings highlight the effectiveness and trustworthiness of the proposed algorithm in securing the communication process and maintaining visual integrity.

关键词： Steganography most significant bit(MSB) least significant bit(LSB) peak signal-to-noise ratio(PSNR)

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：