检索结果-内蒙古大学图书馆

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Lu, Bo-Han Lin, Yi-Hsuan Lee, En-Shiun Annie Tsai, Richard Tzong-Han Department of Computer Science and Information Engineering National Central University Taiwan Department of Computer Science Faculty of Arts and Science University of Toronto Canada Computer Science Program Faculty of Science Ontario Tech University Canada Center for GIS Research Center for Humanities and Social Sciences Academia Sinica Taiwan

ISBN: (纸本)9782493814104

Machine translation focuses mainly on high-resource languages (HRLs), while low-resource languages (LRLs) like Taiwanese Hokkien are relatively under-explored. The study aims to address this gap by developing a dual translation model between Taiwanese Hokkien and both Traditional Mandarin Chinese and English. We employ a pre-trained LLaMA 2-7B model specialized in Traditional Mandarin Chinese to leverage the orthographic similarities between Taiwanese Hokkien Han and Traditional Mandarin Chinese. Our comprehensive experiments involve translation tasks across various writing systems of Taiwanese Hokkien as well as between Taiwanese Hokkien and other HRLs. We find that the use of a limited monolingual corpus still further improves the model's Taiwanese Hokkien capabilities. We then utilize our translation model to standardize all Taiwanese Hokkien writing systems into Hokkien Han, resulting in further performance improvements. Additionally, we introduce an evaluation method incorporating back-translation and GPT-4 to ensure reliable translation quality assessment even for LRLs. The study contributes to narrowing the resource gap for Taiwanese Hokkien and empirically investigates the advantages and limitations of pre-training and fine-tuning based on LLaMA 2. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Enhancing children’s learning experience through alphabet ar-game: a usability and acceptability assessment of augmented reality game-based learning

引用

Multimedia Tools and Applications 2024年 1-23页

作者： Aslam, Fatima Rahman, Hameedur Wahid, Samiya Abdul Wahid, Saira Abdul Ali, Numan Department of Computer Science & amp IT Institute of Southern Punjab Multan Pakistan Department of Computer Games Development Faculty of Computing & ampAI Air University Islamabad44000 Pakistan

In the evolving landscape of early childhood education, the integration of technology has emerged as a significant paradigm shift. This study focuses on this shift, presenting the development and evaluation of the Alphabet AR-Game—an innovative augmented reality (AR) application designed for preschool children. The game creates a dynamic learning environment, revolutionizing alphabet learning through interactive 3D visualizations and audio elements. While existing research explores AR applications for learning purpose, there remains a gap in the literature, concerning a comprehensive AR tool tailored explicitly for preschoolers, combining gamification, pedagogical methodologies, and stakeholder involvement. Our findings showcase the Alphabet AR-Game’s success in integrating technology with traditional learning, achieving remarkable usability and acceptability among children, parents, and teachers. This study introduces a novel model for early childhood education, highlighting the importance of collaboration between home and educational settings to promote comprehensive cognitive development. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Smart Contract Coordinated Privacy Preserving Crowd-Sensing Campaigns 22

Smart Contract Coordinated Privacy Preserving Crowd-Sensing ...

引用

22nd IEEE Consumer Communications and Networking Conference, CCNC 2025

作者： Bedogni, Luca Ferretti, Stefano University of Modena-Reggio Emilia Department of Computer Science and Mathematics Italy University of Bologna Department of Computer Science and Engineering Italy

ISBN: (纸本)9798331508050

Crowd-sensing has emerged as a powerful data retrieval model, enabling diverse applications by leveraging active user participation. However, data availability and privacy concerns pose significant challenges. Traditional methods like data encryption and anonymization, while essential, may not fully address these issues. For instance, in sparsely populated areas, anonymized data can still be traced back to individual users. Additionally, the volume of data generated by users can reveal their identities. To develop credible crowd-sensing systems, data must be anonymized, aggregated and separated into uniformly sized chunks. Furthermore, decentralizing the data management process, rather than relying on a single server, can enhance security and trust. This paper proposes a system utilizing smart contracts and blockchain technologies to manage crowd-sensing campaigns. The smart contract handles user subscriptions, data encryption, and decentralized storage, creating a secure data marketplace. Incentive policies within the smart contract encourage user participation and data diversity. Simulation results confirm the system's viability, highlighting the importance of user participation for data credibility and the impact of geographical data scarcity on rewards. This approach aims to balance data origin and reduce cheating risks. © 2025 IEEE.

关键词： Anonymity

来源：评论

学校读者我要写书评

暂无评论

More than just a name?From magnetic to optical skyrmions and the topology of light

引用

Light(science & Applications) 2025年第1期14卷 26-29页

作者： Jian Chen Andrew Forbes Cheng-Wei Qiu School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai200093China Department of Electrical and Computer Engineering National University of SingaporeSingapore117583Singapore School of Physics University of the WitwatersrandJohannesburgSouth Africa

Topology is usually perceived intrinsically immutable for a given *** argue that optical topologies do not immediately enjoy such ***'optical skyrmions'as an example,we show that they will exhibit varying textures and topological invariants(skyrmion numbers),depending on how to construct the skyrmion vector when projecting from real to parameter *** demonstrate the fragility of optical skyrmions under a ubiquitous scenario-simple reflection off an optical *** topology is not without benefit,but it must not be assumed.

关键词： numbers space. topology

来源：评论

学校读者我要写书评

暂无评论

Wideband Matched Filtering Antennas with Absorptive Resonators

Wideband Matched Filtering Antennas with Absorptive Resonato...

引用

2023 International Applied Computational Electromagnetics Society Symposium, ACES-China 2023

作者： Cui, Jiaqian Zhu, Lei University of Macau Faculty of Science and Technology Department of Electrical and Computer Engineering China

ISBN: (纸本)9781733509657

This paper presents a design method of filtering antennas with wide matching bands to realize the integration of multiple functions such as absorbing, filtering, and radiating, and protect the system as well. In our design, an absorptive stub is employed to absorb the out-of-band signals. As such, a cascaded topology with lossy stubs, lossless stubs, and the resonant antenna is proposed to acquire the absorptive filtering antenna. The characteristic of the absorptive stub, and the response of the equivalent circuit are specifically analyzed here. Finally, an absorptive filtering slot antenna is implemented to prove that the method can be used to achieve a matching bandwidth far beyond that of a conventional resonant antenna. © 2023 Applied Computational Electromagnetics Society (ACES).

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Smart Hybrid Mini-Xception Model and Eyebrow Displacement-Based Stress Level and Emotion Detection System 12th

Smart Hybrid Mini-Xception Model and Eyebrow Displacement-Ba...

引用

12th International Conference on Recent Trends in Computing, ICRTC 2024

作者： Bhuvana, S. Anantha Prabha, P. Deva Priya, M. Christy Jeba Malar, A. Karthick, S. School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore India Department of Computer Science and Engineering PSG Institute of Technology and Applied Research Tamilnadu Coimbatore India Department of Computer Science and Engineering Karpagam College of Engineering Tamilnadu Coimbatore India Department of Information Technology Sri Krishna College of Technology Tamilnadu Coimbatore India Department of Computer Science and Engineering Delhi - NCR Campus SRM Institute of Science and Technology Uttar Pradesh Ghaziabad India

ISBN: (纸本)9789819789450

Stress is a state of mental or emotional strain due to adversative or challenging situations. A human may undergo bad life experiences or events, and it is a significant issue to be dealt in today's society. It comes as a result of family situations, health concerns, work environments and people's behavior among other things. It affects people of various ages, including children, teenagers, middle-aged people and senior citizens. It has serious consequences affecting human body and life. These problems can be addressed by detecting mental stress at an early stage. The main goal of the propounded system is to identify human stress using a pre-trained Mini-Xception model. Additional measures such as eye moments, mouth activity and head motion are employed for determining the level of human stress. FER2013 dataset is used and the system is tested for all kinds of emotions. Human stress level is experimented by capturing video using webcam. From the experimental outcomes, it is evident that the proposed system offers better performance in terms of Precision, Accuracy, Specificity and Sensitivity in contrast to traditional methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Eyebrow displacement FER2013 dataset Mini-Xception model Stress detection

来源：评论

学校读者我要写书评

暂无评论

Modified MobileNetV2 with Convolutional Block Attention Module for Facial Expression Recognition

Modified MobileNetV2 with Convolutional Block Attention Modu...

引用

2023 International Workshop on Artificial Intelligence and Image Processing, IWAIIP 2023

作者： Nafi'an, Muh. Hafidh Ilmi Bachtiar, Fitra Abdurrachman Setiawan, Budi Darma Brawijaya University Computer Science Faculty Informatics Engineering Department Malang Indonesia

ISBN: (纸本)9798350382914

Facial emotion recognition is one of the artificial intelligence implementations used to recognize emotions based on data learned by computers. Unlike humans, who can recognize a person's emotions directly, computers need several trainings procedures conducted by humans to be able to recognize a person's emotions. Previous research has proposed various methods with deep learning and traditional machine learning approaches to classify emotions based on faces. Some studies obtained relatively high accuracy, but on evaluation by cross-validation, the results were much lower than the accuracy obtained. Therefore, this study proposes an approach using a modified MobileNetV2 deep learning architecture in the residual layer by adding a Convolutional Block Attention Module (CBAM) to improve accuracy and data generalization. This experiment uses the Karolinska Directional Emotional Faces (KDEF) public benchmark dataset. Preprocessing is done with image augmentation to enrich the training data by resizing, center cropping, random horizontal flipping, random affine, and random rotation. The proposed method is evaluated by 5-fold cross-validation and compared with MobileNetV2 without modification and state-of-the-art methods. Experimental results show that the proposed model outperform all comparative methods by achieving a 5 -fold cross-validation accuracy of 94.387%. © 2023 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Efficient Method for Detecting Abnormal Growth of Blood Vessels Using Convolutional Neural Network

Efficient Method for Detecting Abnormal Growth of Blood Vess...

引用

2023 International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent engineering, ICECONF 2023

作者： Kumar, A. Deepak Sasipraba, T. Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai600 119 India Department of Computer Science and Engineering St. Joseph s Institute of Technology Chennai600119 India Faculty of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai600119 India

ISBN: (纸本)9798350334357

Diabetic Retinopathy is the main problem in human life because of high sugar levels present in the blood. Various organs get affected due to this reason. The eye is the one of the parts of the human eye which goes to vision problems sometimes blindness. The early stages of detection need to protect the human eye. Existing model uses various methods which do not solve the problem completely. Machine learning based approach introduced for detection of the affected area of eye. The Blood Vessels of the eye get affected as a result bleeding in the eye and excess growth in the eye. Traditional algorithms are not suitable for detecting this growth rate due to the less resolution images. The CNN based model with integrated data sets are used to classify and detect the blood vessel. The high resolution images are used for detecting the location and exact difference in normal vessels. Various algorithms are used with different data sets for making multidimensional analysis. The objective of this method is to identify the heterogeneous vessels from the normal vessel with a high level of accuracy. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Automated approach to predict cerebral stroke based on fuzzy inference and convolutional neural network

引用

Multimedia Tools and Applications 2024年 1-22页

作者： Alrowais, Fadwa Jamjoom, Arwa A. Karamti, Hanen Umer, Muhammad Alsubai, Shtwai Abate, Andrea F. Ashraf, Imran Department of Information Systems College of Computer and Information Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah21589 Saudi Arabia Department of Computer Science The Islamia University of Bahawalpur Bahawalpur Pakistan Department of Computer Science College of Computer Engineering and Sciences Prince Sattam bin Abdulaziz University P.O. Box 151 Al-Kharj11942 Saudi Arabia Department of Computer Science University of Salerno Fisciano84084 Italy Department of Information and Communication Engineering Yeungnam University Gyeongsan38541 Korea Republic of

Cerebral stroke indicates a neurological impairment caused by a localized injury to the central nervous system resulting from a diminished blood supply to the brain. Today, stroke stands as a global menace linked to the premature mortality of millions of people globally. Consequently, it is crucial to simulate how different risk factors impact the incidence of strokes and artificial intelligence is emerging as a suitable tool for achieving this goal. This study aims to construct dependable learning prediction models for stroke illness. The proposed approach can handle the inherent difficulty posed by a significant imbalance in classes, where the number of stroke patients is notably smaller compared to the healthy class. The study also provides a model based on an adaptive neuro-fuzzy inference system logic and convolutional neural networks (CNN) for accurate stroke prediction. An adaptive neuro-fuzzy inference system logic approach is adopted as it incorporates the capabilities of artificial intelligence and fuzzy inference, thereby having the potential to yield superior outcomes. The efficacy of the suggested method is extensively analyzed involving machine and deep learning classifiers and considering metrics relevant to encompass both the capacity for generalization and the accuracy in predicting. The proposed fuzzy-CNN model outperforms with the most considerable accuracy of 98.97% when using the original dataset, resampled dataset, and data imputation. K-fold cross-validation also shows superior results with an average accuracy of 99.6% for five folds. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Plutchik Wheel of Emotion and Machine Learning-Based Hybrid Sentiment Analysis for the Hindi Language with Minimum Dependency on High Computation Resources

引用

SN computer science 2023年第6期4卷 797页

作者： Kumar, Pravin Vardhan, Manu Department of Computer Science and Engineering Nit Raipur Chhattisgarh Raipur India Department of Electronics and Communication (Computer Science and Engineering) University of Allahabad Prayagraj India

Sentiment analysis is a natural language processing technique for extracting sentimental, opinion, or emotional information from text data. Sentiment analysis of natural language is a complex task that requires high computation resources, and Sentiment analysis becomes more complicated if the language of text data is Hindi. This manuscript proposes a plutchik wheel of emotion and Machine learning-based hybrid sentiment analysis approach for the Hindi language with minimum dependency on high computation resources. In the proposed methodology, we assembled a lexicon for the Hindi language. Each word of the lexicon is mapped with respected plutchik emotion. As very few standardised data sets are available for the sentiment analysis of the Hindi language, thus we drafted a new data set. The data set compiled is a collection of educational kid stories from different publication books. Our experiment study shows that the proposed approach gives 73.45% accuracy on this data set with a lexicon comparable to state-of-the-art machine learning and lexicon-based hybrid sentiment analysis approaches. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Hindi Machine learning Natural language processing Sentiment analysis

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：