检索结果-内蒙古大学图书馆

A transformer model for boundary detection in continuous sign language

Multimedia Tools and Applications 2024年第42期83卷 89931-89948页

作者： Rastgoo, Razieh Kiani, Kourosh Escalera, Sergio Electrical and Computer Engineering Faculty Semnan University Semnan*** Iran Sergio Escalera Department of Mathematics and Informatics Universität de Barcelona and Computer Vision Center Barcelona Spain

Sign Language Recognition (SLR) has garnered significant attention from researchers in recent years, particularly the intricate domain of Continuous Sign Language Recognition (CSLR), which presents heightened complexity compared to Isolated Sign Language Recognition (ISLR). One of the prominent challenges in CSLR pertains to accurately detecting the boundaries of isolated signs within a continuous video stream. Additionally, the reliance on handcrafted features in existing models poses a challenge to achieving optimal accuracy. To surmount these challenges, we propose a novel approach utilizing a Transformer-based model. Unlike traditional models, our approach focuses on enhancing accuracy while eliminating the need for handcrafted features. The Transformer model is employed for both ISLR and CSLR. The training process involves using isolated sign videos, where hand keypoint features extracted from the input video are enriched using the Transformer model. Subsequently, these enriched features are forwarded to the final classification layer. The trained model, coupled with a post-processing method, is then applied to detect isolated sign boundaries within continuous sign videos. The evaluation of our model is conducted on two distinct datasets, including both continuous signs and their corresponding isolated signs, demonstrates promising results. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Gesture recognition

来源：评论

学校读者我要写书评

暂无评论

Predicting from a Different Perspective: A Re-ranking Model for Inductive Knowledge Graph Completion 21st

Predicting from a Different Perspective: A Re-ranking Mode...

引用

21st Pacific Rim International Conference on Artificial Intelligence, PRICAI 2024

作者： Iwamoto, Yuki Kaneiwa, Ken Department of Computer Network Engineering Graduate School of Informatics and Engineering The University of Electro-Communications Tokyo Japan

ISBN: (纸本)9789819601240

Rule-induction models have demonstrated great power in the inductive setting of knowledge graph completion. In this setting, the models are tested on a knowledge graph entirely composed of unseen entities. These models learn relation patterns as rules by utilizing subgraphs. Providing the same inputs with different rules leads to differences in the model’s predictions. In this paper, we focus on the behavior of such models. We propose a re-ranking-based model called ReDistLP (Re-ranking with a Distinct Model for Link Prediction). This model enhances the effectiveness of re-ranking by leveraging the difference in the predictions between the initial retriever and the re-ranker. ReDistLP outperforms the state-of-the-art methods in 2 out of 3 benchmarks. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Tile-size aware bitrate allocation for adaptive 360∘ video streaming

引用

Multimedia Tools and Applications 2025年第13期84卷 12615-12632页

作者： Huang, Jiawei Liu, Mingyue Liu, Jingling Gao, Feng Li, Weihe Wang, Jianxin School of Computer Science and Engineering Central South University Hunan Changsha410083 China School of Informatics The University of Edinburgh Edinburgh United Kingdom

360∘ videos have become increasingly popular recently, but consume much more bandwidth than non-360∘ videos. Usually, 360∘ video streaming partitions the video surface into multiple tiles and encodes the tiles independently to effectively and flexibly use limited link bandwidth. However, current bitrate adaptive algorithms generally aim to maximize the bitrate, rather than perceptual quality, resulting in degradation of user experience. More importantly, we reveal that the distribution of tile size is very skewed, that is, a small number of large tiles consumes more bandwidth than a large number of small tiles, further hurting the overall viewing quality. Therefore, in this paper, we propose a tile-size aware bitrate allocation scheme TSA for adaptive 360∘ video streaming to improve the viewing experience of users. Specifically, TSA cautiously decreases the quality of a few large tiles to allocate more bandwidth to large number of small tiles, thus improving the perceptual quality of overall video, without sacrificing large tiles excessively. Experiments over real-world datasets show that TSA effectively improves V-VMAF by up to 39% compared with several state-of-the-art adaptive bitrate algorithms. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Video streaming

来源：评论

学校读者我要写书评

暂无评论

Advanced CNN Architectures for Improved Garbage Image Classification: An In-depth Evaluation 9

Advanced CNN Architectures for Improved Garbage Image Classi...

引用

9th South-East Europe Design Automation, computer engineering, computer Networks and Social Media Conference, SEEDA-CECNSM 2024

作者： Kanavos, Athanasios Papadimitriou, Orestis Vonitsanos, Gerasimos Maragoudakis, Manolis Mylonas, Phivos University of the Aegean Department of Information and Communication Systems Engineering Samos Greece University of Patras Computer Engineering and Informatics Department Patras Greece Ionian University Department of Informatics Corfu Greece University of West Attica Department of Informatics and Computer Engineering Athens Greece

ISBN: (纸本)9798350342482

Convolutional Neural Networks (CNNs) have become instrumental in advancing image classification, particularly in the context of garbage image classification, a critical component for efficient waste management. This paper introduces a tailored CNN architecture that demonstrates enhanced accuracy in garbage classification tasks, even with constrained datasets. Our architecture incorporates multiple convolutional, max-pooling, and fully connected layers, with dropout regularization strategically applied to curb overfitting and improve model generalization across a varied waste image dataset. Comparative evaluations reveal that our model achieves a significant improvement in accuracy over existing CNN models. The results not only validate the robustness of our approach but also contribute valuable insights toward developing more precise and efficient systems for garbage image classification. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Advancing Sentiment Analysis of IMDB Movie Reviews with a Hybrid Multinomial Naive Bayes and LSTM Approach 4th

Advancing Sentiment Analysis of IMDB Movie Reviews with a...

引用

4th International Conference on Novel and Intelligent Digital Systems, NiDS 2024

作者： Papadimitriou, Orestis Kanavos, Athanasios Vonitsanos, Gerasimos Maragoudakis, Manolis Mylonas, Phivos Department of Information and Communication Systems Engineering University of the Aegean Samos Greece Computer Engineering and Informatics Department University of Patras Patras Greece Department of Informatics Ionian University Corfu Greece Department of Informatics and Computer Engineering University of West Attica Athens Greece

ISBN: (纸本)9783031733437

Natural Language Processing (NLP) is increasingly pivotal in the natural sciences, with sentiment analysis emerging as a crucial application in the era of big data. Efficiently and accurately extracting meaningful insights from extensive textual data remains a significant challenge. This study explores sentiment analysis within the context of movie reviews, a domain where understanding nuanced viewer perceptions can influence industry outcomes. Traditional methods often struggle with the complexity of language, where identical phrases may convey different sentiments based on their context. To address these challenges, this research implements a hybrid approach, combining a Multinomial Naive Bayes Classifier and a Long Short-Term Memory (LSTM) model developed in TensorFlow. Our integrated methodology not only advances the robustness of sentiment analysis tools but also achieves a notable accuracy of 96.25%, underscoring its potential to enhance complex NLP tasks in real-world applications. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Vision Transformer for Audio-Based Depression Detection on Multi-Lingual Audio Data 24

Vision Transformer for Audio-Based Depression Detection on M...

引用

7th International Conference on Digital Medicine and Image Processing, DMIP 2024

作者： Pratiwi, Monica Sanjaya, Samuel Ady Department of Computer Engineering Faculty of Engineering and Informatics Universitas Multimedia Nusantara Tangerang Indonesia Department of Information System Faculty of Engineering and Informatics Universitas Multimedia Nusantara Tangerang Indonesia

ISBN: (纸本)9798400709586

Depression has the potential to impact death rates, particularly when it comes to death by suicide. Inadequate diagnosis may result in a delay or unsuitable therapy, which can worsen symptoms of depression. Unaddressed or insufficiently addressed depression can result in deteriorating mental well-being, which includes a higher risk of experiencing suicidal ideation and engaging in self-destructive actions. Voice analysis can be employed to distinguish between individuals with depression and those without depression. However, the research that has been conducted on voice recognition to distinguish between depressed and non-depressed individuals keeps focusing on the use of a single dataset source as input for the classification model that is being developed. This work utilizes the Vision Transformer model and various pre-trained transformers, such as DeiT (Data-Efficient Image Transformer) and Swin Transformer, to detect depression based on voice. The objective was to create a more generic model not restricted to a specific language. Integrating Mel-spectrogram characteristics with a vision transformer-based model can enhance the efficacy of voice recognition models when dealing with multi-language data. The result shows 21% higher accuracy than the previous study that also implemented a cross-dataset test. Copyright © 2024 held by the owner/author(s). Publication rights licensed to ACM.

关键词： Spectrographs

来源：评论

学校读者我要写书评

暂无评论

Enhancing Emotion Classification with a Hybrid BERT and CNN Architecture 19

Enhancing Emotion Classification with a Hybrid BERT and CNN ...

引用

19th International Workshop on Semantic and Social Media Adaptation and Personalization, SMAP 2024

作者： Papadimitriou, Orestis Kanavos, Athanasios Vonitsanos, Gerasimos Maragoudakis, Manolis Karkazis, Panagiotis Mylonas, Phivos University of the Aegean Department of Information and Communication Systems Engineering Samos Greece University of Patras Computer Engineering and Informatics Department Patras Greece Ionian University Department of Informatics Corfu Greece University of West Attica Department of Informatics and Computer Engineering Athens Greece

ISBN: (纸本)9798331504502

In computational linguistics, effectively encoding and systematizing emotional expressions in language is a significant challenge. Existing machine learning (ML) models for text analysis often only recognize primary emotional states such as anger, happiness, and sadness, missing the more nuanced spectrum of human emotions. These models frequently overlook the complex semantic interplays and fail to capture the full diversity of emotional expressions, focusing instead on simplistic categorization. This paper proposes a robust ML framework that enhances emotion classification by discriminating among nuanced emotion categories - Anger, Joy, and Fear. Leveraging a sophisticated combination of Convolutional Neural Networks (CNNs) and the Bidirectional Encoder Representations from Transformers (BERT) architecture, our framework demonstrates exceptional accuracy in emotion detection. Evaluating a diverse dataset of text samples, each meticulously tagged with its expressed emotion, confirms the model's superior performance in recognizing and classifying a broad range of emotional states. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Security Strategy of Digital Medical Contents Based on Blockchain in Generative AI Model

引用

computers, Materials & Continua 2025年第1期82卷 259-278页

作者： Hoon Ko Marek R.Ogiela Division of Computer Science and Engineering Sunmoon UniversityAsan31460Republic of Korea AGH University of Krakow Cryptography and Cognitive Informatics LaboratoryKrakow30-059Poland

This study presents an innovative approach to enhancing the security of visual medical data in the generative AI environment through the integration of blockchain *** combining the strengths of blockchain and generative AI,the research team aimed to address the timely challenge of safeguarding visual medical *** participating researchers conducted a comprehensive analysis,examining the vulnerabilities of medical AI services,personal information protection issues,and overall security *** multi faceted exploration led to an indepth evaluation of the model’s performance and ***,the correlation between accuracy,detection rate,and error rate was *** analysis revealed insights into the model’s strengths and limitations,while the consideration of standard deviation shed light on the model’s stability and performance *** study proposed practical improvements,emphasizing the reduction of false negatives to enhance detection rate and leveraging blockchain technology to ensure visual data integrity in medical *** blockchain to generative AI-created medical content addresses key personal information protection *** utilizing the distributed ledger system of blockchain,the research team aimed to protect the privacy and integrity of medical data especially medical *** approach not only enhances security but also enables transparent and tamperproof ***,the use of generative AI models ensures the creation of novel medical content without compromising personal information,further safeguarding patient *** conclusion,this study showcases the potential of blockchain-based solutions in the medical field,particularly in securing sensitive medical data and protecting patient *** proposed approach,combining blockchain and generative AI,offers a promising direction toward more robust and secure medical content *** research and advancements

关键词： Digitalmedical content medical diagnostic visualization security analysis generativeAI blockchain vulnerability pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Quantum-Based Prediction Model for Carbon Neutrality 11

Quantum-Based Prediction Model for Carbon Neutrality

引用

11th International Conference on Advanced informatics: Concept, Theory and Application, ICAICTA 2024

作者： Munawar, Ghifari Surendro, Kridanto School of Electrical Engineering and Informatics Institut Teknologi Bandung Bandung Indonesia Politeknik Negeri Bandung Department of Computer Engineering and Informatics Bandung Indonesia

ISBN: (纸本)9798331520311

Carbon neutrality is a global target pursued by cities worldwide to achieve a balance between carbon emissions and removals, reaching a net-zero carbon state. Mitigation measures are being implemented to reduce emissions and enhance carbon sequestration, aiming to meet the targets set for 2050 or 2060. However, challenges posed by urban sprawl and increasing urbanization raise concerns about the feasibility of achieving carbon neutrality. Various studies have been conducted to project the attainment of this goal by developing prediction models. Machine learning (ML) prediction models use socio-economic, energy, and technological data to forecast carbon neutrality. These models consider factors like GDP per capita, urbanization rate, total energy consumption, and forest stock volume, formulating scenarios based on policy documents and historical data. Some models have incorporated optimization methods like the sparrow search algorithm, genetic neural network, and aquila optimizer to improve prediction accuracy. However, classical optimization methods have limitations, such as susceptibility to getting trapped in local optima, which can affect model performance. Quantum-based optimization methods, particularly quantum annealing (QA), are emerging as potential solutions to address these challenges by leveraging the principles of quantum mechanics to optimize complex problem spaces. QA enhances ML processes like feature selection, hyperparameter optimization, and regression model optimization. This study provides a review of pipeline processes from state-of-the-art methods, as well as their potential quantum-based enhancements, to achieve more precise predictive models. © 2024 IEEE.

关键词： Carbon sequestration

来源：评论

学校读者我要写书评

暂无评论

Bias analysis of AI models for undergraduate student admissions

引用

Neural Computing and Applications 2025年第12期37卷 7785-7795页

作者： Van Busum, Kelly Fang, Shiaofen Department of Computer Science and Software Engineering Butler University IndianapolisIN United States Computing and Engineering Indiana University Luddy School of Informatics Indiana University Indianapolis IndianapolisIN United States

Bias detection and mitigation is an active area of research in machine learning. This work extends previous research done by the authors Van Busum and Fang (Proceedings of the 38th ACM/SIGAPP Symposium on Applied Computing, 2023) to provide a rigorous and more complete analysis of the bias found in AI predictive models. Admissions data spanning six years was used to create an AI model to determine whether a given student would be directly admitted into the School of Science under various scenarios at a large urban research university. During this time, submission of standardized test scores as part of a student’s application became optional which led to interesting questions about the impact of standardized test scores on admission decisions. We developed and analyzed AI models to understand which variables are important in admissions decisions, and how the decision to exclude test scores affects the demographics of the students who are admitted. We then evaluated the predictive models to detect and analyze biases these models may carry with respect to three variables chosen to represent sensitive populations: gender, race, and whether a student was the first in his/her family to attend college. We also extended our analysis to show that the biases detected were persistent. Finally, we included several fairness metrics in our analysis and discussed the uses and limitations of these metrics. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Students

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：