检索结果-内蒙古大学图书馆

Self-supervised rotation-aware guided attentive distilled transformer network for crowd face detection

Neural Computing and applications 2025年 1-27页

作者： Jayanthan, K.S. Domnic, S. Department of Computer Applications National Institute of Technology Tamil Nadu Tiruchirappalli620015 India

Accurate crowd face detection is vital in surveillance, security, and crowd management due to challenges like varying densities, occlusions, and scale variations. Resizing issues arise when traditional techniques fail to handle diverse face sizes and densities, alongside patterns or objects resembling faces, hindering accuracy. To address this, we propose the Self-Supervised Rotation-aware Guided Attentive Distilled Transformer Network (SSRGADTN) for crowd face detection. This method combines self-supervised learning and a rotation-aware distilled transformer network, effectively utilizing unlabeled data to learn robust representations and handle diverse crowd scenes. The workflow starts with self-supervised training on the FWL dataset, followed by integration with the Guided Attentive Detector Network (GADN) trained on the WIDERFACE dataset to enhance face detection, forming the SSRGADN (Self-Supervised Rotation-aware Guided Attentive Detector Network) model. Knowledge distillation further refines the model using a transformer network as the student model trained on the CrowdHuman dataset, resulting in the SSRGADTN model, which outperforms other state-of-the-art models in crowd face detection accuracy. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Classifying the state of knowledge-based question answering: patterns, progress, and prospects

引用

International Journal of computers and applications 2025年第1期47卷 93-105页

作者： Sharma, Sumit Jain, Sarika Tiwari, Mukesh Kumar lal, Sohan Department of Computer Science and Engineering Quantum University Roorkee Uttarakhand Roorkee India Department of Computer Applications National Institute of Technology Haryana Kurukshetra India

In response to inquiries posed in natural languages, question-answering systems (QASs) produce responses. The capabilities of early QASs are limited because they were designed for certain domains. The current generation of QASs focuses on the types of inquiries that users typically enquire, the features of the data sources that are used, and the formats in which the right responses are produced. Many QASs have been developed since the 1960s when research on the subject first started. There are several cutting-edge frameworks for answering questions that can be used to categorize the QA system according to a number of criteria. The necessity for an extensive survey on QASs comes naturally in order to determine the future focus of research in this field. In this study, QASs are analyzed by the number of criteria used to categorize question-answering (QA) and provide the dimension using a variety of standards to cover all the QAS. We describe the current state of research in each QAS area and make recommendations for its future direction. Based on the dimension criteria outlined in the research methodology, research studies are chosen for additional review and comparison. Constructed some criteria to classify the QASs, we provide the classification and go over the benefits and drawbacks of each class's QAS, as well as the problems with the corresponding study. Results are detailed summarized and compared. As a result of this study, a general architecture of frameworks for QA over knowledge base (KB) is drawn in addition to the comparative analysis of systems, serving as a reference for academics working on related topics. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Progressive reasoning: a cognitively inspired method for interesting pattern extraction in cognitive IoT

引用

Iran Journal of computer Science 2025年 1-18页

作者： Jha, Vidyapati Tripathi, Priyanka Department of Computer Applications National Institute of Technology Chhattisgarh Raipur India

This research pioneers the integration of cognitive capabilities into the Internet of Things (IoT), creating a transformative subfield known as cognitive IoT (CIoT). Building on IoT’s foundational principles, CIoT addresses the immense challenge of deriving meaningful insights from the vast data it generates. To tackle this complexity, the study proposes progressive reasoning, a cognitively inspired method for processing sensory data and uncovering significant patterns. At its core, progressive reasoning treats sensory observations as relationships, seeking their transitive closure—a state that signifies the progression of data as it transitions through various states. This progression enables the transformation of raw data into meaningful insights, revealing patterns and their probabilistic occurrences. By quantifying and analyzing this transformation, the research offers a robust mechanism to extract value from CIoT datasets. The methodology’s effectiveness is validated through experimental analysis of environmental and healthcare data, achieving an impressive accuracy of over 99.26%. This success highlights its potential to revolutionize critical applications in fields like healthcare and environmental monitoring by enabling smarter, more efficient systems. This study not only advances the technical boundaries of CIoT but also underscores the power of blending cognitive processes with technology to drive deeper understanding and impactful innovation. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Cognitive IoT Heterogeneous Massive Progressive reasoning

来源：评论

学校读者我要写书评

暂无评论

Fine-Tuned T5 Transformer with LSTM and Spider Monkey Optimizer for Redundancy Reduction in Automatic Question Generation

引用

SN computer Science 2024年第5期5卷 475页

作者： Tharaniya sairaj, R. Balasundaram, S.R. Department of Computer Applications National Institute of Technology Tiruchirappalli India

The significance of Automatic Question Generation (AQG) lies in its potential to support educators and streamline assessment processes. Notable improvements in AQG are seen with the use of language models, ranging from LSTMs to Transformers. However, there is a need for improvement in the probabilistic scoring technique employed for next word generation in the target question. In this regard, it is noted that template-based methods offer potential for enhancement, although they may result in the generation of fewer or redundant questions due to the utilization of fixed templates. This research aims to address this gap by proposing a hybrid model that combines the the advantages of template-based and Transformer-based AQG approaches. The template-based LSTM approach is explored to learn adaptable question templates. On the other side, the Transformer model is explored to reduce redundancy in the auto-generated questions. The proposed work finetunes the pipelined T5 Transformer model using the Spider Monkey Optimizer over the LSTM-generated templates. The choice of Spider Monkey Optimizer enhances the selection of the named entity in question tail (tail entity) through dynamic sub-search space division for efficient exploration and exploitation, and self-organization based on local and global scoring. This ensures that the named entity in the question tail is non-redundant (diverse) and regards both structural and contextual coherence to the auto-generated question. Experimental findings highlight improvements in how well the diversely selected named entities are relevant to the generated questions through higher precision, recall and f1-scores in the pipelining phase. Moreover, the study shows that the Spider Monkey Optimizer performs better in selecting tail entities, and it consistently outperforms other algorithms in F1-score and convergence time across all datasets, with its time complexity increasing linearly as dataset size grows. The finetuned pipelined T5 mod

关键词： Automatic question generation Generative fine-tuning LSTM Spider monkey optimizer Transformer model

来源：评论

学校读者我要写书评

暂无评论

Imprecise reasoning: extracting precision in imprecision for knowledge discovery in cognitive IoT

引用

Iran Journal of computer Science 2025年第2期8卷 607-625页

作者： Jha, Vidyapati Tripathi, Priyanka Department of Computer Applications National Institute of Technology Chhattisgarh Raipur India

An emerging field within the internet of things (IoT) known as cognitive IoT (CIoT) has gained attention due to the growing interest in integrating cognitive capabilities into IoT systems. Similar to IoT, which consists of a network of interconnected devices, CIoT generates vast amounts of data that are diverse, unpredictable, and time-sensitive. Consequently, efficiently extracting computational knowledge from these large volumes of real-time sensor data presents a significant challenge for CIoT. This study proposes a novel approach to identifying meaningful concepts and patterns through imprecise reasoning. Initially, the proposed method addresses the impact of noisy inputs, followed by the construction of a concept lattice. The weight of the generated concepts is computed using plausibility values, allowing for the selection of minimally weighted (significant) concepts. In the next phase, relevant data are extracted from these significant concepts, which are then combined to form meaningful patterns. The effectiveness of the proposed approach is empirically evaluated using weather data spanning 21.25 years. A variety of metrics were employed to validate the proposed method, and the results, with an accuracy exceeding 99.57%, demonstrate that the proposed algorithm outperforms existing competitors. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Cognitive IoT Concept lattice Imprecise reasoning Significant pattern

来源：评论

学校读者我要写书评

暂无评论

Adaptive E-Learning Environments: A Methodological Approach to Identifying and Integrating Multi-layered Learning Styles

引用

SN computer Science 2024年第6期5卷 772页

作者： Ramesh, M. Jayashree, R. Department of Computer Science and Applications Faculty of Science and Humanities SRM Institute of Science and Technology Department of Computer Applications Faculty of Science and Humanities SRM Institute of Science and Technology

E-learning environments represent digital platforms designed to facilitate online learning experiences. Recognizing the diverse learning preferences of individuals, the need for identifying and integrating multi-layered learning styles within these environments is paramount. Existing approaches often face limitations in accurately capturing and accommodating these intricate learning styles. To address these challenges, this study proposes a novel approach. Firstly, the integration of word embedding-based feature extraction transforms textual data into continuous vector representations, enhancing feature robustness and meaningfulness. Secondly, leveraging deep belief networks (DBNs) allows for the automatic learning of hierarchical data representations, improving model performance by capturing complex patterns. Thirdly, the DBN-based model accommodates multi-layered learning styles, including visual, auditory, kinaesthetic, and read/write preferences, offering personalized recommendations to learners. Lastly, the proposed approach is scalable and generalizable, capable of handling large datasets and diverse educational contexts, thus enhancing the efficacy of e-learning platforms. This innovative approach demonstrates promising advancements in learning style identification within e-learning environments, providing personalized guidance to learners and improving the overall effectiveness of online education. The proposed Deep Belief Network (DBN) model exhibits a significant average increase in accuracy compared to other methods implemented in Python. With an accuracy of 99.5%, the DBN model surpasses the accuracy of the K-Nearest Neighbours (KNN) and Random Forest (RF) models by 25.4 and 3.3%, respectively. This substantial improvement underscores the superiority of the proposed DBN model in learning style identification. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Deep belief networks E-learning environments Learning style identification Multi-layered learning styles Word embedding

来源：评论

学校读者我要写书评

暂无评论

Nonlinear system dynamics of calcium and nitric oxide due to cell memory and superdiffusion in neurons

引用

Communications in Theoretical Physics 2024年第5期76卷 16-32页

作者： Anand Pawar Kamal Raj Pardasani Department of Mathematics Bioinformatics and Computer ApplicationsMaulana Azad National Institute of TechnologyBhopal-462003Madhya PradeshIndia

The integer-order interdependent calcium([Ca^(2+)])and nitric oxide(NO)systems are unable to shed light on the influences of the superdiffusion and memory in triggering Brownian motion(BM)in ***,a mathematical model is constructed for the fractional-order nonlinear spatiotemporal systems of[Ca^(2+)]and NO incorporating reaction-diffusion equations in *** two-way feedback process between[Ca^(2+)]and NO systems through calcium feedback on NO production and NO feedback on calcium through cyclic guanosine monophosphate(cGMP)with plasmalemmal[Ca^(2+)]-ATPase(PMCA)was incorporated in the *** Crank–Nicholson scheme(CNS)with Grunwald approximation along spatial derivatives and L1 scheme along temporal derivatives with Gauss–Seidel(GS)iterations were *** numerical outcomes were analyzed to get insights into superdiffusion,buffer,and memory exhibiting BM of[Ca^(2+)]and NO *** conditions,events and mechanisms leading to dysfunctions in calcium and NO systems and causing different diseases like Parkinson’s were explored in neurons.

关键词： [Ca^(2+)]and nitric oxide signaling reaction-diffusion equations fractional-order systems Grunwald approximation Gauss-Seidel iterations

来源：评论

学校读者我要写书评

暂无评论

Voice Authentication at Scale: A Room Level Approach for Smart Devices

Voice Authentication at Scale: A Room Level Approach for Sma...

引用

2025 IEEE International Students' Conference on Electrical, Electronics and computer Science, SCEECS 2025

作者： Dharmalingam, Anandhakumar Tirumala, P. Department of Information Technology & Computer Applications Guntur India

ISBN: (纸本)9798331529833

In light of the rapid expansion of voice- activated smart devices, achieving reliable and secure voice authentication across multiple rooms presents complex challenges. Traditional voice authentication methods fall short in differentiating speakers based on their specific locations, raising security risks in shared spaces. This study introduces a room-level voice authentication model that leverages both spatial audio features and machine learning techniques to distinguish users based on their physical environment. By analyzing room-specific acoustic signatures and integrating them with biometric voice characteristics, the proposed approach strengthens access control, offering a scalable solution that addresses the unique demands of multi- room setups. Performance evaluations indicate high accuracy in correctly identifying both speaker and room, confirming the method's potential for secure, location-based voice authentication in smart homes and office spaces. © 2025 IEEE.

关键词： Smart homes

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence for Climate Sustainability: A Comprehensive Review of applications, Challenges, and Future Prospects 4

Artificial Intelligence for Climate Sustainability: A Compre...

引用

4th International Conference on Sustainable Expert Systems, ICSES 2024

作者： Parameshwari, M. Gnanaguru, Gnaneswari Vtu Department of Computer Applications Belagavi India Cmr Institute of Technology Department of Computer Applications Bengaluru India

ISBN: (数字)9798331540364

ISBN: (纸本)9798331540364

Climate sustainability plays a very crucial role in environmental sustainability and by considerable use of natural resources, and reducing pollution, we can make our planet beautiful and sustaining for the life of future generations. Climate changes constitute increasingly severe challenges, and with the use of Artificial Intelligence (AI), the necessity for innovation in fostering climate sustainability has become increasingly evident. This review organizes an exhaustive inventory of which AI affects climate science. The challenges remain unmet in AI's various applications across sectors such as optimization of renewable energy, climate modeling, disaster resilience and environmental monitoring. This paper discusses the important roles of applying AI in overcoming climate change difficulties. These include energy management or conservation, cap-and-trade for carbon emissions, climate and weather forecasting, grid management, transportation, and agriculture infrastructure, or city planning. It also looks into how and the extent to which AI is adopted in resource management, increasing the adoption of renewable energy, fostering sustainable development, and overseeing resources such as water and forests. This study acknowledges various barriers that impede the effective application of Artificial Intelligence in combating climate change which include data restrictions, algorithms' bias, and ethical issues with AI operations. To solve the major challenges described, we improve the major findings of this extensive literature review and distill a set of practical recommendations for policy-makers, industrial actors and researchers. Which will in return assist in overcoming challenges and enhance the capacity of AI in combating climate change and environmental conservation. © 2024 IEEE.

关键词： Zero-carbon

来源：评论

学校读者我要写书评

暂无评论

Video captioning – a survey

引用

Multimedia Tools and applications 2025年第2期84卷 947-978页

作者： Vaishnavi, J. Narmatha, V. Department of Computer Applications SRM Institute of Science and Technology Trichy India Department of Computer and Information Science Annamalai University Annamalai Nagar Tamilnadu 608002 India

The explosion of the novel phenomenon of the combination of computer vision and Natural language processing is playing a vital role in converting the ordinary world into a more technological pool. Natural language processing and computer vision are the vanguard of Artificial intelligence with the enormous potential currently ruling more fields. The trending Accelerated research of the NLP and CV combo are cancer screening, surgical simulation, visual properties description, visual retrieval, visual description, etc. Video captioning is a task of localizing the event and various features are extracted from the particular event and then the description of the entire event is performed with Natural languages using various techniques. The task of video captioning has some evolution of various techniques from Traditional methods to Deep learning. Currently, Deep learning techniques are ruling the field. Several models are being proposed for the better enhancement of the video captioning task with the default Encoder-Decoder framework and some models with the variance of Attention mechanism and Transfer learning etc., The Attainment of every technique and the accuracy of the generated results are highly dependent on the nature of the problem, diversity, performance of chosen dataset and construction of the technique with various layers, and the amount of data split for train, validate, and test. This survey discusses the captioning types, real-time applications of video captioning, evolution of various methods in video captioning, advanced techniques, different datasets, metrics that are used to evaluate the results, and analysis of the results of the existing models for video captioning. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： computer vision

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：