检索结果-内蒙古大学图书馆

2024 International Symposium on Computing and Intelligent Systems, SCI 2024

作者： Chaurasia, Anushka Kumar, Deepak Yogita Department of Computer Science and Engineering National Institute of Technology Meghalaya India Department of Computer and Engineering National Institute of Technology Kurukshetra India

Neurological adverse drug reactions (NADRs) pose a significant clinical challenge as they can have a profound impact on patient health and treatment outcomes. While diverse drug descriptors have been employed for neurological ADR prediction, the potential of using 17 molecular descriptors for this purpose has not been explored. To address this, a multilabel NADR-Net and MLSMOTE-based framework have been proposed for neurological adverse drug reaction prediction. The data for 17 MD and ADRs were extracted from PubChem and ADRECs databases and then mapped based on drug ID. The resulting dataset contained information on 2160 drugs, including their molecular properties, and 1030 ADRs. The methodology was then applied to this dataset, and it showed promising results in terms of hamming loss, precision, true positive rate, f1 score, and ROC-AUC. This study highlights the potential of using molecular descriptors for predicting neurological ADRs, which could improve patient outcomes and drug safety. © 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Word recognition from Indian Sign Language in videos using dual feature descriptor and GMT-MASKRCNN recognition technique

引用

Multimedia Tools and Applications 2025年第5期84卷 2565-2597页

作者： Bansal, Naman Jain, Abhilasha Computer Science and Engineering Department Maharaja Ranjit Singh Punjab Technical University Bathinda India

Indian Sign Language (ISL) is a vital means of interaction for the country's deaf and hard-of-hearing citizens. This highly developed visual-gestural system of language uses forms, gestures with the hands, and facial reactions to convey an extensive variety of complex ideas and viewpoints. Nevertheless, the automated ISL movement recognition technique has a great problem because of the varying characteristics and intricacy of sign lingo sentences. In order to overcome that, the researchers present a thorough approach to the recognition of Indian Sign Language (ISL), utilizing a multi-phase analytic framework that combines sophisticated signal processing techniques with machine learning to improve the accuracy and robustness of ISL gesture detection. The model employs the Central Limits Theorem-Discrete Wavelet Transform (CLT-DWT) eliminating strategies to reduce interference from noise in recorded sign language videos. It also improves the stage of pre-processing by using FICA-KMC for overlapping pixel division and concentrates on resilient extraction of features through the Manhattan Distance-based Histogram of Oriented Gradients (MD-HOG) and CCC-SURF strategies. Furthermore, researchers incorporate MIC-DT for choosing features in order to optimize the detection model's effectiveness. To achieve comprehensive motion recognition and categorization, the final procedure is to incorporate the powerful artificial neural network architecture GMT-Mask RCNN. The technique's possible efficacy in precisely and consistently comprehending Indian Sign Language for the visually impaired belonging is highlighted by the exploratory evaluation, which shows significant improvements in the rate of recognition as well as effective prevention of erroneous positive and negative incidents. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Performance Comparison of Machine Learning Methods for Flood Prediction 9th

Performance Comparison of Machine Learning Methods for Flood...

引用

9th International Conference on computer science and Computational Intelligence, ICCSCI 2024

作者： Linus, Josel Tanjaya, Vincent Diana Kurniawan, Afdhal Computer Science Department School of Computer Science Bina Nusantara University Jakarta14480 Indonesia

Flood is a disaster that causes problems. It causes problems such as destruction of property and disruption of civil activities. Predicting floods is a solution to prevent the devastating disaster and is crucial for mitigation. This study shows a comparison of various machine learning classification methods for flood prediction using a dataset consisting of humidity, wind speed, wind direction, location data, and other related features. We evaluated the performance of logistic regression, Naïve Bayes, K-Nearest Neighbors (KNN), decision tree, and random forest models based on metrics such as accuracy, precision, recall, and the area under the receiver operating characteristic curve (ROC-AUC). The results indicate that random forest achieved the highest accuracy at 95.902%, followed by KNN with 95.130%, decision tree with 91.448%, logistic regression with 79.513%, and naïve bayes with 77.256%. This comparative analysis highlights the strengths and limitations of each model, providing valuable insights for selecting the most appropriate approach for flood prediction. The results highlight the role of machine learning in creating an effective way of predicting floods. © 2024 The Authors.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Industrialized Implementation of Agile Methodology 1

Industrialized Implementation of Agile Methodology

引用

1st IEEE International Conference on Advances in Computing, Communication and Networking, ICAC2N 2024

作者： Singh, Meeta Nandal, Poonam Bura, Deepa Jindal, Srishty School of Engineering and Technology Department of Computer Science and Engineering Faridabad India

ISBN: (纸本)9798350356816

This paper introduces a guide aimed at aligning academic learning with industry standards in Agile methodology and Scrum framework usage. Targeting students, educators, and industry practitioners, it addresses the prevalent gap between theoretical knowledge and practical *** with an overview of Agile's iterative approach, it delves into Scrum's key elements: sprint planning, daily stand-ups, reviews, and retrospectives. It acknowledges the disconnect between academic curricula and industry demands, especially in Agile contexts. © 2024 IEEE.

关键词： Agile manufacturing systems

来源：评论

学校读者我要写书评

暂无评论

Secure Information Sharing with Access Control and Tamper Detection for Drone-Based Disaster Management 13

Secure Information Sharing with Access Control and Tamper De...

引用

13th IEEE Global Conference on Consumer Electronic, GCCE 2024

作者： Kawahara, Yuki Fujisawa, Masaya Mohd Kamal, Ahmad Akmal Aminuddin Tokyo University of Science Department of Information and Computer Technology Tokyo Japan

ISBN: (纸本)9798350355079

In a disaster, reliable and secure information exchange between affected individuals and disaster management centers is essential to facilitate rescue activities. However, in large-scale disasters, a method with low computational cost is preferable for reducing the communication and computation volume. In this study, we propose a cost-effective information-sharing system with access control and tamper detection. © 2024 IEEE.

关键词： Disaster prevention

来源：评论

学校读者我要写书评

暂无评论

A Comparison of BiLSTM, BERT, and Ensemble Method for Emotion Recognition on Indonesian Product Reviews 9th

A Comparison of BiLSTM, BERT, and Ensemble Method for Emotio...

引用

9th International Conference on computer science and Computational Intelligence, ICCSCI 2024

作者： Pramana, Rio Jonathan, Marcel Yani, Habel Steven Sutoyo, Rhio Computer Science Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia

Emotion recognition within online product reviews is pivotal for enhancing business strategies and customer insights. Recognizing emotions in Indonesian, a language rich in nuances and expressions presents significant challenges due to its complex linguistic structure and the scarcity of tailored datasets. This study aims to bridge this gap by evaluating the effectiveness of Bidirectional Long Short-Term Memory (BiLSTM), Bidirectional Encoder Representations from Transformers (BERT), and ensemble methods in analyzing emotions from Indonesian product reviews using the PRDECT-ID dataset. Extensive fine-tuning across 23 BERT configurations and multiple BiLSTM preprocessing combinations was conducted to adapt these models to the Indonesian linguistic context. Each model was assessed based on the F1 score. The BiLSTM model was particularly effective in configurations with complex preprocessing, achieving an optimal F1 score of 61% through advanced noise removal, stemming, and a modified stopwords list. Conversely, the minimally preprocessed fine-tuned'base-p2' and'large-p1' BERT variants achieved F1 scores of 72% and 73%, respectively, both surpassing the previous best result of 71%. This research also explored 4 ensemble methods, combining the strengths of the best-performing BiLSTM and BERT models using both soft-voting and stacked generalization techniques. The unweighted stacked generalization achieved a 74% F1 score, while the weighted method excelled with the highest F1 score of 75%, surpassing other models and highlighting the advantages of strategic model integration. This research significantly advances the development of NLP models for Indonesian text, demonstrating how tailored deep-learning approaches can effectively enhance emotion recognition accuracy. © 2024 The Authors.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Human identification system based on ear shape using convolutional neural network 9th

Human identification system based on ear shape using convolu...

引用

9th International Conference on computer science and Computational Intelligence, ICCSCI 2024

作者： Antonio, Randy Tyandra, Nadya Indryani, Tiffany Angela Edbert, Ivan Sebastian Aulia, Alvina Computer Science Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia

Convolutional Neural Networks (CNN) represents a class of Deep Neural Networks frequently employed for image and video recognition, image classification, medical image analysis, and numerous computer vision tasks. In human identification systems, fingerprint and facial recognition are widely used. However, fingerprints are subject to deterioration, and both face and finger could become less recognizable in the presence of a cut or a wound. The ear, a less-known component in the identification system, eliminates the less advantageous traits of fingerprint and facial recognition. The methodology employed encompasses the systematic collection of datasets, model development, and model evaluation. Evaluation procedures involve user assessments conducted objectively about findings from analogous studies. The EarVN1.0 dataset serves as the foundation, and five distinct models, namely MobileNetV3-Large, VGGNet-19, ResNet50V2, EfficientNetV2-Small, and a modified version of AlexNet, have been meticulously developed. The outcomes of this research are delineated by the accuracies achieved by each model, which stand at 79%, 22%, 79%, 91%, and 52%, respectively. An in-depth exploration involving hyperparameter tuning of the EfficientNetV2-S model, focusing on key aspects such as batch size, pooling layer, dropout, and maximum epoch limit. This meticulous tuning resulted in a notable accuracy improvement, elevating the performance of the EfficientNetV2-S model to 92%. © 2024 The Authors.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Human Activity Recognition in Construction Industry Using Machine Learning Pose Estimation Technique

Human Activity Recognition in Construction Industry Using Ma...

引用

International Conference on Interdisciplinary Approaches in Civil Engineering for Sustainable Development, IACESD 2023

作者： Manoj Kumar, M. Hegde, Bhuvaneshwari Veda Murthy, S.P. Akhila, M.K. Bhoomika, A.S. Department of Computer Science and Engineering Jyothy Institute of Technology Bengaluru India

ISBN: (纸本)9789819996094

In many applications, including measuring physical activity, understanding sign language, and controlling full-body gestures, human position estimation from video is essential. This has the potential to be utilized for activity recognition in civil work. By accurately tracking human body posture from video, the technology can identify and classify different tasks and actions being performed by workers in construction, manufacturing, or other industries. This information can be used to monitor worker productivity, optimize workflow, and identify potential safety hazards. The proposed project is a machine learning (ML) solution for high-fidelity body posture tracking, employing current open source research that also drives the Machine Learning Pose Detection Application programming interface to infer predefined 3D landmarks and background segmentation mask on the entire body from RGB (Red, Green, Blue) video frames. The suggested method in this project achieves real-time performance on the majority of modern mobile phones, desktops/laptops, Python, and even the web, in contrast to current state-of-the-art methodologies, which rely mostly on strong desktop environments for inference. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Cross-lingual Editing in Multilingual Language Models 18

Cross-lingual Editing in Multilingual Language Models

引用

18th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2024 - Findings of EACL 2024

作者： Beniwal, Himanshu Kowsik Nandagopan, D. Singh, Mayank Department of Computer Science and Engineering Indian Institute of Technology Gandhinagar India

ISBN: (纸本)9798891760936

The training of large language models (LLMs) necessitates substantial data and computational resources, and updating outdated LLMs entails significant efforts and resources. While numerous model editing techniques (METs) have emerged to efficiently update model outputs without retraining, their effectiveness in multilingual LLMs, where knowledge is stored in diverse languages, remains an underexplored research area. This research paper introduces the cross-lingual model editing (XME) paradigm, wherein a fact is edited in one language, and the subsequent update propagation is observed across other languages. To investigate the XME paradigm, we conducted experiments using BLOOM, mBERT, and XLM-RoBERTa using the two writing scripts: Latin (English, French, and Spanish) and Indic (Hindi, Gujarati, and Bengali). The results reveal notable performance limitations of state-of-the-art METs under the XME setting, mainly when the languages involved belong to two distinct script families. These findings highlight the need for further research and development of XME techniques to address these challenges. For more comprehensive information, the dataset used in this research and the associated code are publicly available at the following URL. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

A New Approach for Prevention of Malicious Attack in Post Disaster Using Delay Tolerant Network

A New Approach for Prevention of Malicious Attack in Post Di...

引用

2024 International Conference on Intelligent Systems for Cybersecurity, ISCS 2024

作者： Chakrabarti, Chandrima Bal, Saptarshi Jaiswal, Rishabh Ansari, Arbaz Akhtar Narula Institute of Technology Department of Computer Science and Engineering Kolkata India

ISBN: (纸本)9798350375237

Natural disasters still have a crippling effect on our infrastructure notwithstanding society's development. Communication systems, the backbone of our network, sustain the most damage during such crises. When it comes to setting up working communication systems, volunteers working on disaster relief missions face enormous obstacles because of the deteriorating communication infrastructure and extended power outages. Smartphones have become essential tools because they may create a peer-to-peer network, in opportunistic way also called as a Delay Tolerant Network (DTN), through Bluetooth or Wi-Fi connections in a latency mode, which makes communication easier in post-disaster situations. Nevertheless, nodes in this network can act strangely, failing to create routes or send data to other nodes, which would hinder the functioning of the network as a whole. Malicious nodes are those that behave in this way. They have the ability to either deny or destroy messages, which can cause serious disruptions. The performance of the network may be further complicated by some nodes displaying partial malicious activity, which causes interference on a selective basis. In response to these difficulties, this work suggests a novel technique for identifying and mitigating the effects of malicious nodes coordinating assaults inside DTNs. Through simulations with the Opportunistic Network Environment (ONE) simulator, the efficacy of the scheme was assessed, demonstrating its ability to reduce the disruptive impact of malicious nodes on network operations. This suggested method shows promise in strengthening DTNs against malevolent attacks and promoting more dependable communication infrastructures in places affected by disasters. © 2024 IEEE.

关键词： Delay tolerant networks

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：