检索结果-内蒙古大学图书馆

5th European Tangible Interaction Studio, ETIS 2022

作者： Echeverri, Daniel Department of Visual Computing Faculty of Informatics Masaryk University Brno Czech Republic

This paper expands on previous work on tangible narratives. It briefly reviews the design case of Letters to José, from which stemmed a narrative architecture for tangible narratives and a typology that frames the qualities of tangible artifacts with narrative meaning. Based on this architecture and exemplified in the typology, the paper introduces The Non-myth of the Noble Red, a tangible narrative which expands upon the learnings gained from the design of Letters to José. This paper presents some of its technological considerations and discusses how it integrates Brechtian theatre concepts to explore alternative digital storytelling approaches. © 2022 Copyright for this paper by its authors.

关键词： Brechtian theatre interactive storytelling tangible interaction Tangible Narratives The Nonmyth of the Noble Red

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Review of Bias Detection and XAI Techniques in NLP 5

A Comprehensive Review of Bias Detection and XAI Techniques ...

引用

5th International Conference on Advanced Research in computing, ICARC 2025

作者： Bhavaneetharan, Loganathan Informatics Institute of Technology Department of Computing Colombo Sri Lanka

ISBN: (纸本)9798331530983

Bias in Natural Language Processing (NLP) models pose urgent ethical and societal challenges by perpetuating stereotypes and inequalities. This review provides a comprehensive overview of state-of-the-art Explainable AI (XAI) techniques - such as LIME, SHAP, and Integrated Gradients - for detecting and mitigating bias. We detail our methodology for selecting relevant studies and analyze key NLP datasets (e.g., StereoSet, CrowS-Pairs) to uncover specific limitations like language coverage and intersectional gaps. Moreover, we highlight emerging challenges in scalability, cultural diversity, and regulatory compliance. By integrating these findings, we propose actionable strategies for fair and transparent NLP systems, positioning this work as a foundation for ongoing improvements in equitable AI solutions. © 2025 IEEE.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Self-Paced Partial Domain-Aware Learning for Face Anti-Spoofing 48

Self-Paced Partial Domain-Aware Learning for Face Anti-Spoof...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Chen, Zhiyi Lu, Yao Deng, Xinzhe Meng, Jia Zhang, Shengchuan Cao, Liujuan Xiamen University Media Analytics and Computing Lab Department of Artificial Intelligence School of Informatics China Youtu Lab Tencent China

ISBN: (纸本)9781728163277

With the widespread deployment of face authentication systems, domain generalization (DG) based face anti-spoofing (FAS) security approaches have drawn growing attention. Existing generalization-based methods always attempt to extract domain-invariant task information from data and eliminate domain-dependent information from representation space. However, they neglect that domain-related information may also contain helpful features for the classification task. To address this issue, we propose a self-paced partial domain-aware framework (SPDA) to preserve domain-related features helpful for the discrimination of fake and real faces, thereby increasing generalization for unseen domains. Specifically, a training strategy based on contrastive learning is adopted to construct domain-adapted and domain-aware task-related representation spaces. Then, a partial domain-aware adaptation module (PDA) is proposed to preserve valuable domain-related information for the task features that the network considers useful for mixture-domain classification. In addition, the proposed self-paced method(SCM) continuously explores potential clusters with insufficient representation to enhance further the feature extractor's capability and the effectiveness of the PDA module. Extensive experiments demonstrate the effectiveness of our method compared to SOTA algorithms. © 2023 IEEE.

关键词： Contrastive Learning Domain Generalization Face Anti-spoofing

来源：评论

学校读者我要写书评

暂无评论

FracturePro: Automated Bone Fracture Identification and Classification Using an Ensemble Deep Learning Approach 5

FracturePro: Automated Bone Fracture Identification and Clas...

引用

5th International Conference on Advanced Research in computing, ICARC 2025

作者： Rodrigo, Madushani Athuraliya, Banuka Informatics Institue of Technology Department of Computing Colombo Sri Lanka

ISBN: (纸本)9798331530983

In modern medical diagnostics, machine learning (ML) has emerged as a transformative tool, reshaping healthcare practices. This study investigates the application of ML for the timely and accurate detection of bone fractures, a critical aspect of orthopedic care. Delayed diagnosis can result in complications such as malunion or non-union, underscoring the need for real-time, precise identification of fracture types and locations. While X-ray interpretation relies heavily on the expertise of medical professionals, poor image quality often poses significant challenges. To address these issues, this research introduces a novel ML-based model for fracture segmentation and classification. The model leverages advanced architectures, including ResNet, DenseNet, and U-Net, to perform fracture segmentation and classification with high accuracy. Additionally, image enhancement and preprocessing techniques are incorporated to mitigate the limitations posed by low-quality radiographic images. The findings highlight the potential of ML in improving diagnostic precision and efficiency, ultimately enhancing patient outcomes in orthopedic care. © 2025 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

FracturePro: Automated Bone Fracture Identification and Classification Using an Ensemble Deep Learning Approach

FracturePro: Automated Bone Fracture Identification and Clas...

引用

Advanced Research in computing (ICARC), International Conference on

作者： Madushani Rodrigo Banuka Athuraliya Department of Computing Informatics Institue of Technology Colombo Sri Lanka

ISBN: (数字)9798331530983

ISBN: (纸本)9798331530990

关键词： Location awareness Image quality Image segmentation Accuracy Spirals Computational modeling Computer architecture Bones Diagnostic radiography X-ray imaging

来源：评论

学校读者我要写书评

暂无评论

Early fire danger monitoring system in smart cities using optimization-based deep learning techniques with artificial intelligence

引用

Journal of Reliable Intelligent Environments 2024年第2期10卷 197-210页

作者： Reddy, P. Dileep Kumar Margala, Martin Shankar, S. Siva Chakrabarti, Prasun Telangana Secunderabad India School of Computing and Informatics University of Louisiana at Lafayette Lafayette United States Department of CSE KG Reddy College of Engineering and Technology Telangana Hyderabad501504 India Department of Computer Science and Engineering Sir Padampat Singhania University Rajasthan Udaipur313601 India

One primary safety concern for smart cities is fire. Traditional techniques are not appropriate because of their high false alarm rates, delayed characteristics, and susceptibility in situations with heritage buildings. Smart cities must develop sophisticated methods to mitigate the severe effects of fires and achieve early fire detection in real time. An artificial intelligence-based recurrent neural network with a whale optimization framework (AI-RNN-WO) was introduced to estimate the risk of fire hazards early on. IoT sensor devices are first deployed in smart cities to continuously monitor environmental parameters such as temperature, smoke, flame, relative humidity, fuel moisture, and duff moisture code. These sensed data are then saved in the cloud storage system Firebase. Then, the sensed dataset is updated to the designed model, which pre-processes the data and extracts relevant features from the dataset. The RNN parameters are tuned using whale optimization, which improves the prediction results and attains better accuracy. The performance of the proposed AI-RNN-WO model is validated using a MATLAB tool, and the performance is compared with existing models. The produced model has demonstrated its effectiveness by attaining the highest accuracy (99.5%) and lowest error rate (0.1%). © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Job Resumes Recommendation using Integration of Fuzzy Discernibility Matrix Feature Selection and Convolutional Neural Network Multi-label Text Classification

Informatica (Slovenia)

引用

Informatica (Slovenia) 2025年第13期49卷 49-58页

作者： Hayami, Regiolina Yusoff, Nooraini Daud, Kauthar Mohd Fatma, Yulia Department of Informatics Engineering Faculty Computer Science Universitas Muhammadiyah Riau Indonesia Faculty of Data Science and Computing Universiti Malaysia Kelantan 16100 Malaysia Center for Artificial Intelligence Technology Faculty of Information Science and Technology Universiti Kebangsaan Malaysia 43600 Malaysia

The manual analysis of job resumes poses specific challenges, including the time-intensive process and the high likelihood of human error, emphasizing the need for automation in content-based recommendations. Recent advancements in deep learning, particularly Convolutional Neural Networks (CNN) for Multi-label Text Classification (MLTC), offer a promising solution for addressing these challenges through artificial intelligence. While CNN is renowned for its robust feature extraction capabilities, it faces specific challenges such as managing high-dimensional data and poor data interpretability. To address these limitations, this study employs the Fuzzy Discernibility Matrix (FDM) feature selection technique to determine the relevance of skills for each job vacancy. FDM assigns weights to the features of each job category and ranks their relevance based on the highest scores, which are then utilized in the MLTC CNN model. The integration of FDM and MLTC CNN serves as the foundation for generating content-based recommendations derived from key features in job resumes. This study produces a content-based job recommendation system that displays the top three job categories, accompanied by explanations of the skills supporting each selected category. These recommendations also consider cosine similarity values between analyzed items and the integrated results of FDM and MLTC CNN. The application of FDM for feature weighting has been proven to enhance multi-label text classification outcomes by providing better insights during the feature selection process. With a recall of 97.26%, precision of 94.81%, and accuracy of 98.58%, the MLTC model integrating FDM feature selection and CNN demonstrates robust performance characteristics in content-based job recommendation tasks. © 2025 Slovene Society Informatika. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning-based Optimization of EBike Charging infrastructure 11

Reinforcement Learning-based Optimization of EBike Charging ...

引用

11th IEEE/ACM International Conference on Big Data computing, Applications and Technologies, BDCAT 2024

作者： Sharif, Muddsair Seker, Huseyin Birmingham City University Faculty of Computing Engineering and the Built Environment BirminghamB5 5JU United Kingdom The University of Sharjah College of Computing and Informatics Department of Information Systems Sharjah United Arab Emirates

ISBN: (纸本)9798350367300

This paper presents a novel technique for enhancing the allocation of resources in the charging infrastructure for e-bikes by employing deep reinforcement learning (DRL) in a context-specific manner. With the evolving transportation land-scape and the increasing popularity of electric micro-mobility solutions, efficient resource management plays a vital role. The methodology involves utilizing deep Q networks (DQN) in virtual environments to dynamically allocate charging resources based on contextual parameters. Through empirical analysis, we demonstrate the efficacy of our approach in achieving optimal resource allocation while considering the needs of diverse stakeholders. Findings suggest that the DQN model, when integrated with context-awareness and stakeholder objectives, enables efficient resource allocation, maximize fleet utilization, ensures charging station availability, enhances user satisfaction, and promotes environmental sustainability. The findings of this study offer valuable insights for policymakers, urban planners, and industry stakeholders seeking to enhance the sustainability and effectiveness of micro-mobility services in urban settings. Overall, this study contributes to the advancement of sustainable urban mobility solutions and underscores the potential of DRL in addressing complex urban transportation challenges. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Aspect-Based Sentiment Analysis of Student Reviews on MOOC Platforms 5

Aspect-Based Sentiment Analysis of Student Reviews on MOOC P...

引用

5th International Conference on Advanced Research in computing, ICARC 2025

作者： Dissanayake, A.D.M.B.B. Fernando, Pumudu A. Informatics Institute of Technology Department of Computing Colombo Sri Lanka

ISBN: (纸本)9798331530983

The rapid growth of Massive Open Online Courses (MOOCs), particularly in the post-COVID-19 era, has transformed education and led to a substantial increase in student-generated reviews. However, manual analysis of these reviews is impractical due to their sheer volume, and traditional binary sentiment analysis lacks the granularity needed to extract actionable insights. This study proposes a novel framework leveraging Aspect-Based Sentiment Analysis (ABSA) to classify MOOC reviews into seven critical aspects. The model was trained on a 100,000-review dataset from Coursera and a manually annotated 10,000-review dataset from Udemy. Various machine learning techniques were evaluated, with the final model integrating Long Short-Term Memory (LSTM) networks and Support Vector Machines (SVM). The SVM achieved notable performance, with accuracy rates of 93.7% for sentiment prediction and 88.64% for aspect classification on the Coursera dataset, and 73.91% for sentiment prediction and 66.95% for aspect classification on the Udemy dataset. These results underscore the model's efficacy in delivering a comprehensive and platform-independent approach to analyzing MOOC reviews, providing valuable insights for educators and course developers. © 2025 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Analysis of the Effectiveness of a New Composite Method Joining Features of Implicit and Modular Enumeration Algorithms

Analysis of the Effectiveness of a New Composite Method Join...

引用

2022 International Russian Automation Conference, RusAutoCon 2022

作者： Groppen, Vitaliy Oskarovich Berko, Aleksey Aleksandrovich North Caucasian Institute of Mining and Metallurgy Department of Informatics and Computing Vladikavkaz Russia

ISBN: (数字)9781665466554

ISBN: (纸本)9781665466554

A new method of searching for globally optimal solutions of discrete programming problems is proposed, which combines the components of search strategies by modular enumeration, backtracking, and dynamic programming. Analytical and experimental results of the analysis of its comparative efficiency in relation to backtracking search solving the knapsack problem are presented. It has been experimentally proven that, unlike the modular enumeration method and one of its varieties - the 'meet in the middle' algorithm, the optimal number of modules of the proposed approach is not a constant, it increases with the number of variables of the problem being solved. Its high operational characteristics have been experimentally confirmed: a relatively low solution search time and low requirements for the amount of RAM. In the general case, we can say that the higher efficiency of this approach compared to the exhaustive or implicit enumeration methods increases with the complexity of the problem it solves. © 2022 IEEE.

关键词： Dynamic programming

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：