检索结果-内蒙古大学图书馆

2022 Findings of the Association for Computational Linguistics: NAACL 2022

作者： James, Jesin Yogarajan, Vithya Shields, Isabella Watson, Catherine Keegan, Peter J. Jones, Peter-Lucas Mahelona, Keoni Dept. of Electrical Computer and Software Engineering The University of Auckland New Zealand Strong AI Lab School of Computer Science The University of Auckland New Zealand Te Puna Wananga The University of Auckland New Zealand Te Hiku Media New Zealand

ISBN: (纸本)9781955917766

Te reo Maori, New Zealand's only indigenous language, is code-switched with English. Maori speakers are atleast bilingual, and the use of Maori is increasing in New Zealand English. Unfortunately, due to the minimal availability of resources, including digital data, Maori is under-represented in technological advances. Cloud-based multilingual systems such as Google and Microsoft Azure support Maori language detection. However, we provide experimental evidence to show that the accuracy of such systems is low when detecting Maori. Hence, with the support of Maori community, we collect Maori and bilingual data to use natural language processing (NLP) to improve Maori language detection. We train bilingual sub-word embeddings and provide evidence to show that our bilingual embeddings improve overall accuracy compared to the publiclyavailable monolingual embeddings. This improvement has been verified for various NLP tasks using three bilingual databases containing formal transcripts and informal social media data. We also show that BiLSTM with pretrained Maori-English sub-word embeddings outperforms large-scale contextual language models such as BERT on down streaming tasks of detecting Maori language. However, this research uses large models 'as is' for transfer learning, where no further training was done on Maori-English data. The best accuracy of 87% was obtained using BiLSTM with bilingual embeddings to detect Maori-English codeswitching points. © Findings of the Association for Computational Linguistics: NAACL 2022 - Findings.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Virtualization of Control Systems in IoT using Neural Networks 13

Virtualization of Control Systems in IoT using Neural Networ...

引用

13th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops, ICUMT 2021

作者： Slepnev, Aleksey Karelin, Evgenii Luksha, Ivan Kirichek, Ruslan Zadorozhnyaya, Alina Makarov, Eugene Bonch-Bruevich Saint Petersburg State University of Telecommunications Dept. of Software Engineering and Computer Science Saint-Petersburg Russia

ISBN: (纸本)9781665402194

In the last few years, there has been a huge leap in the development of the Internet and data transmission networks in general. The number of Internet of Things devices is increasing every year, and by 2030 their estimated number will reach 125 billion. In accordance with this, the issue of traffic optimization and the transition to a new network architecture is becoming more and more relevant. Here is a method for optimization industrial traffic using a software and hardware complex for aggregation and prediction of commands by a neural network in this paper. Training based on artificially simulated data shows the result of at least 90% of the predictions of the entire chain of commands from the server to the client. and copy;2021 IEEE. © 2021 IEEE computer Society. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

From Alert to Change: Effects of Sensing on Construction Workers’ Health Protection Behaviours

SSRN

引用

SSRN 2024年

作者： Zhou, Na Chang-Richards, Alice Dirks, Kim Natasha Wang, Kevin I-Kai Dept. of Civil and Environmental Engineering Faculty of Engineering the University of Auckland New Zealand Engineering Block 5 - Bldg. 405 5 Grafton Rd Auckland Central Auckland1010 New Zealand Engineering Block 1 - Bldg. 401 20 Symonds St Auckland Central Auckland1010 New Zealand Dept. of Electrical Computer and Software Engineering Faculty of Engineering the University of Auckland New Zealand

Compromised air quality on a construction site could lead to significant adverse respiratory impacts on workers. This paper introduces an alert sensing methodology to develop practical health and wellbeing requirements based on the Health Belief Model (HBM)Health Belief Model:(HBM) and the Theory of Planned Behaviour (TPB)Theory of Planned Behaviour:(TPB). The approach leverages the Internet of Things (IoT)Internet of Things:(IoT) to enable real-time, data-driven evaluation of air quality metrics, facilitating the assessment of workers' behavioural change. Dust sensors were mounted on the helmets of 11 construction workers at two New Zealand sites for 21 days, and pre and post-questionnaires were conducted to gather feedback on their experience before and after using the system. The research findings revealed that sensing systems can significantly increase awareness and positive behavioural change amongst construction workers with respect to particulate matter (PM)particulate matter:(PM) exposure. This research expands the HBM and the TPB by demonstrating how perceived health benefits, perceived barriers, self-efficacy, perceived behavioural control, and attitudes influence construction workers' behavioural intentions toward the sensing system. The case study also highlights the potential of sensing technologies to foster healthier behaviours and workplace environments, offering insights for construction companies to develop comprehensive health and wellbeing programs that reduce PM exposure risks and promote proactive worker actions. © 2024, The Authors. All rights reserved.

关键词： Air quality

来源：评论

学校读者我要写书评

暂无评论

A Study on Traceable Group Signature Scheme for Blockchain

A Study on Traceable Group Signature Scheme for Blockchain

引用

Information and Network Technologies (ICINT), International Conference on

作者： Hyo-Jin Song Im-Yeong Lee Dept. of Software Convergence Soonchunhyang Univ. Asan-si 31538 Republic of Korea Dept. of Computer Software Engineering Soonchunhyang Univ. Asan-si 31538 Republic of Korea

A blockchain is a decentralized database system in which a ledger containing transactions is stored and maintained identically by all through consensus among network participants without a central server. Instead of having a central server, network participants make sure that everyone stores and maintains the same ledger containing transactions through consensus. Integrity and reliability of the data stored in the blockchain are guaranteed, but on the other hand, privacy issues may arise. The pseudonymity of the public key reused in transactions causes a privacy invasion of the user’s behavioral pattern, asset size, and offline identity. Therefore, instead of assigning unique public keys to everyone, they are provided through a single signature in a blockchain environment. It can also be described as a method for ensuring anonymity by having members of the same group use the same public key in a transaction while employing the group signature, which has been studied in this paper. This paper defines the key used for group signature is defined as a composition of a “group private key” and a “partial key”. Mainly among them, we consider the leakage of the partial key. In the proposed scheme, it is possible to trace the leaker and the signer from the signature generated by the outsider with the leaked partial key. In this paper, we provide a secure traceable group signature scheme to cope with the leakage of the group signing key.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PV-Cap: 3D Dynamic Scene Understanding Through Open Physics-based Vocabulary

PV-Cap: 3D Dynamic Scene Understanding Through Open Physics-...

引用

IEEE computer Society Conference on computer Vision and Pattern Recognition Workshops (CVPRW)

作者： Hidetomo Sakaino Thao Nguyen Phuong Vinh Nguyen Duy Data Solution Dept. AI-Image Group (FCJ.ABC) FPT Consulting Japan FPT Software Department of Computer Science and Communications Engineering Waseda University

ISBN: (数字)9798350365474

ISBN: (纸本)9798350365481

Recently, large Vision Language (VL) models, i.e., CLIP, have demonstrated impressive capabilities in training solely on internet-scale image-language pairs. Moreover, almost all VL models have tackled indoor objects under controlled illumination and camera views. However, outdoor 3D environments are time-varying uncontrolled scenes under natural phenomena. Therefore, captions from such unseen scenes and objects are hard to obtain in a state-of-the-art (SOTA) one-shot algorithm, resulting in insufficient captions. This paper proposes PV-Cap (Physics-based Vocabulary for Caption) for enhancing 3D scene understanding through enriched captions. Since many tasks in understanding 3D dynamic scenes are hard to deal with, PVCap aims to disentangle such complexities through multiple grouped Deep Learning and Vision Language models step-wisely. Proposed i-VQA (iterative VQA) and 3D-CPP (3D Contrastive Physical-Scale Pretraining) extended from SOTA 2D-CLIP also contribute to generating physical and 3D-based captions. Using many images with 3D dynamic events, i.e., road scenes with traffic flow and accidents, experiments have demonstrated the usability and effectiveness of proposed PV-Cap over SOTA models in terms of segmentation and captions.

关键词： Deep learning Training Solid modeling Vocabulary Three-dimensional displays Roads Cameras

来源：评论

学校读者我要写书评

暂无评论

A systematic literature review on generating higher-order mutant 8

A systematic literature review on generating higher-order mu...

引用

8th International Conference on Computing for Sustainable Global Development, INDIACom 2021

作者： Amare, Melashu Mohapatra, Sudhir Kumar Panigrahy, Tarini Prasad Addis Ababa Science Technology University Dept. of Software Engineering Addis Ababa Ethiopia Gandhi Institute for Technological Advancement Dept. of Computer Sc. and Engineering Bhubaneswar India

ISBN: (纸本)9789380544434

Mutation testing is a fault-based testing technique that is used to evaluate the quality of a given set of test cases by inserting a single change into the program under test and then checking whether tests are good enough to detect the artificial injected defects or not. The objective of mutation testing is not testing the actual program rather it is focusing on measuring the effectiveness of test cases by using muted code to check whether the test suite is capable to identify the real fault. This paper summarizes and analyzes the current knowledge about higher-order mutants through a systematic literature review. This Systematic Literature Review analyzes the results of more than one hundred articles in the area of generating higher order mutant. The article presents qualitative results for the surveyed articles. In addition, it augments these results with scientific findings and quantitative results from the primary literature. As a result of this work, this SLR presents an outline for many future works. © 2021 Bharati Vidyapeeth, New Delhi. Copy Right in Bulk will be transferred to IEEE by Bharati Vidyapeeth.

关键词： Testing

来源：评论

学校读者我要写书评

暂无评论

SPORTS CENTER CUSTOMER SEGMENTATION: A CASE STUDY

arXiv

引用

arXiv 2024年

作者： Soto, Juan Carmenaty, Ramón Lastra, Miguel Fernández-Luna, Juan Manuel Benítez, José M. Computer Science and Artificial Intelligence Dept. DiCITS Lab iMUDS Universidad de Granada Spain Software Engineering Dept. DiCITS Lab iMUDS Universidad de Granada Spain

Customer segmentation is a fundamental process to develop effective marketing strategies, personalize customer experience and boost their retention and loyalty. This problem has been widely addressed in the scientific literature, yet no definitive solution for every case is available. A specific case study characterized by several individualizing features is thoroughly analyzed and discussed in this paper. Because of the case properties a robust and innovative approach to both data handling and analytical processes is required. The study led to a sound proposal for customer segmentation. The highlights of the proposal include a convenient data partition to decompose the problem, an adaptive distance function definition and its optimization through genetic algorithms. These comprehensive data handling strategies not only enhance the dataset reliability for segmentation analysis but also support the operational efficiency and marketing strategies of sports centers, ultimately improving the customer experience. © 2024, CC BY-NC-ND.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

A New Compensation Topology For Inductive Power Transfer (IPT) Systems

A New Compensation Topology For Inductive Power Transfer (IP...

引用

Annual Conference of Industrial Electronics Society

作者： You-Chun Huang Udaya Madawala Craig A. Baguley Yao-Ching Hsieh Dept. of Electrical Engineering National Sun Yat-sen University Kaohsiung Taiwan Dept. of Electrical Computer and Software Engineering University of Auckland Auckland New Zealand Dept. of Electrical and Electronic Engineering Auckland University of Technology Auckland New Zealand

This paper presents a new IPT compensation topology that can be operated at two different frequencies to form either primary series and secondary parallel (S-P) or primary parallel and secondary parallel (P-P) compensation. Model are presented to demonstrate how this transformation between S-P and P-P compensations with dual frequency control. The simulated results of a 1 kW IPT system have verified that the proposed system can remain output power with phase-shift modulation (PSM) and allows for efficient power transfer despite large pad misalignments by operating at different compensation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effective Disease Recognition in Cucumbers: A Web-Based Application Using Transfer Learning Models

Effective Disease Recognition in Cucumbers: A Web-Based Appl...

引用

Robotics, Automation, Artificial-intelligence and Internet-of-Things (RAAICON), 2019 IEEE International Conference on

作者： Md Shahidur Rahman Asif Ahamed Md Nuruzzaman Pranto Md Ariful Islam Abdullah Al Masum Abdullah Al-Sakib S M Arafat Rahman Rezaul Haque Shafiur Rahman Dept. of Engineering Management Westcliff University CA USA College of Technology and Engineering Westcliff University CA USA Dept. of Business Administration International American University CA USA Dept. of Information Technology Westcliff University CA USA Dept. of Computer Science and Engineering East West University Dhaka Bangladesh Dept. of Software Engineering Daffodil International University Dhaka Bangladesh

ISBN: (数字)9798331534400

ISBN: (纸本)9798331534417

Cucumber farming plays a crucial role in Bangladesh's agricultural economy, significantly contributing to vegetable production. However, diseases like Downy Mildew, Bacterial Wilt, Anthracnose, and Belly Rot threaten crop quality and lead to financial losses. Effective management of these diseases is crucial for maintaining high crop yields and ensuring economic stability. Current disease identification methods are labor-intensive, costly, and dependent on expert knowledge. To address these challenges, this study presents an automated disease categorization system using advanced transfer learning models, including MobileNet, ShuffleNet, InceptionResNet-v2, AlexNet, and Ensemble Transfer Learning (ETL). Utilizing a dataset of 8,000 well-annotated images, enhanced through preprocessing techniques like GLRM feature extraction and InfoMAX-GAN feature selection, the ETL model achieved a remarkable accuracy of 99.79% and an F1-score of 99.78%. We also developed a user-friendly web tool for real-time diagnostics, providing invaluable support to farmers and agricultural specialists. This tool enables rapid disease detection, leading to increased productivity and a transformative approach to cucumber disease management. By reducing the reliance on chemical interventions, it promotes sustainable farming practices and safeguards human and animal health.

关键词： Deep learning Accuracy Animals Biological system modeling Transfer learning Feature extraction Real-time systems Chemicals Diseases Farming

来源：评论

学校读者我要写书评

暂无评论

AI Assisted Diagnosis of Dengue Amid Arboviral Diseases, Severity Prediction, and Post Effects

AI Assisted Diagnosis of Dengue Amid Arboviral Diseases, Sev...

引用

International Conference on Information Technology Research (ICITR)

作者： Nivasheni B Fernando V.G.S.O. Srihari M Sarah M.I.L. Vishan Jayasinghearachchi Samadhi Chathuranga Rathnayake Dept. of Computer Science Sri Lanka Institute of Information Technology Malabe Sri Lanka Dept. of Software Engineering Sri Lanka Institute of Information Technology Malabe Sri Lanka

ISBN: (数字)9798331511425

ISBN: (纸本)9798331511432

This paper aims to examine the applicability of machine learning models for aiding in the diagnosis and management of arboviral illnesses, especially Dengue. The study focuses on three key objectives: Diagnosis of Dengue, prediction of prognosis of severe disease, and the identification of the post dengue effects. Multiple strategies were used to address these challenges. They are feature engineering based multiclass classification of arboviral diseases, enhancement of standard conventional machine learning algorithms for binary classification of severe Dengue and finally MLSOL enhancing ensemble model for drawing future effects of Dengue infections. The comparison of traditional models included both Random Forest, XGBoost, and also ensemble method while applying the feature selection methods of SFA and RFE. This study presents the comparison of traditional individual models and optimal ensemble method for the multiclass classification in arboviral diseases. In the case of binomial classification, the best accuracy level was 0.93 achieved by the Random Forest with RFE feature selection compared to Logistic Regression as well as the Gradient Boosting. Furthermore, the study proposed an MLSOL augmented ensemble approach to handle label imbalance problem in the dataset which in turn substantially enhanced the prediction accuracy. This approach decreased Hamming Loss to 0. 12 and increased the F-measure to 0.82, this has the capability of handling the imbalanced dataset and improving the predictive accuracy and performance. This machine learning framework offers great potential for clinical use in such aspects as early intervention and in effects following Dengue fever. It is meant to help healthcare practitioners and policy makers on how to better prevent and combat Dengue in Sri Lanka.

关键词： Accuracy Machine learning algorithms Medical services Feature extraction Ensemble learning Object recognition Prognostics and health management Viruses (medical) Random forests Standards

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：