检索结果-内蒙古大学图书馆

data Intelligence and Cognitive Informatics (ICDICI), International Conference on

作者： Talat Binti Firdou Mohammed Ashra A. V. Kalpan Department of Data Science and Business Systems School of Computing SRM Institute of Science and Technology Kattankulathur Tamil Nadu

ISBN: (数字)9798350389609

ISBN: (纸本)9798350389616

Emotion detection is a critical field in AI as well as computer vision, seeking to recognize or detect human emotions from facial expressions. Although this area has been greatly advanced through Convolutional Neural Networks (CNNs), current methods using only single-modal data often exhibit reduced accuracy. In this work, we confront these challenges with a hybrid deep learning method for disparate modalities of information such as speech, video, and text. We pre-process and then use CNN feature extractors for the FER2013 data set. The hybrid deep learning model, hence proposed, achieved a weighted accuracy of 79.3% on FER2013, overcoming the conventional unimodal methods. We are using transformers library with the pre-trained RoBERTa model for text. For speech emotion features, we are using TESS dataset then using a CNN and LSTM (Long Short-Term Memory) model. The CNN extracts feature from the speech, while the LSTM is used for temporal analysis to detect emotions. Once these multimodal features have been obtained, our work would be to fuse in order to get the final emotion recognition using Deep Neural Networks (DNNs). The process increases the precision and reliability of the recognition systems, helping to accumulate even more detailed intuition for different emotional states for a combined approach.

关键词： Deep learning Emotion recognition Accuracy Virtual assistants Speech recognition Feature extraction Convolutional neural networks Reliability Speech processing Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Development and Testing of Mobile Assisted Language Learning Application to Improve Oral Clinical Case Presentation of Student Nurses

International Journal of Interactive Mobile Technologies

引用

International Journal of Interactive Mobile Technologies 2023年第15期17卷 171-188页

作者： Milton, Cynthia Subramaniam, Aruna Sridevi, S. Kumar, S. Ganesh Faculty of Allied Health Sciences Sri Ramachandra Institute of Higher Education Tamil Nadu Chennai India Faculty of Nursing Sri Ramachandra Institute of Higher Education Tamil Nadu Chennai India Sri Ramachandra Engineering and Technology Sri Ramachandra Institute of Higher Education Tamil Nadu Chennai India Department of Data Science and Business Systems School of Computing SRM Institute of Science and Technology Kattankulathur Tamil Nadu Chennai India

Background: Nurses’ oral case presentation is a core clinical communication skill with distinctive linguistic qualities that require specific language training. The present paper is a user feedback analysis of a start-up Mobile Assisted Language Learning application (MALL app) developed and implemented to provide English language training to improve Oral Case Presentation (OCP) of student nurses pursuing baccalaureate degrees from a nursing college in South India. Objective: The aim was to develop and test an Android based OCP-MALL application for user acceptability and effectiveness in improving OCP performance and decreasing the perceived language difficulty. Methodology: Sixty-two student nurses from a reputed nursing college in Chennai, South India, using Android mobile phones participated in the study. All the participants were given access and instructions on the use of OCP-MALL app. A comparison of pre and post-test of perceived level of language difficulty and OCP pre and post score along with users’ acceptance were collected. Results: The users’ acceptability showed that all participants rated the clarity in reading the text with highest mean score of 8.29 and std of 1.276, simplicity in navigation – moving to next page with next highest mean score of 8.19 and std of 1.526, clarity in text application layout with a mean score of 8.14 and std of 1.497 and user friendliness of the application with a mean score of 8.13 and std of 1.393. Conclusion: This startup OCP-MALL app user satisfaction feedback proves that Indian ELT trainers can create cost-effective and need-based m-learning applications. © 2023 by the authors of this article. Published under CC-BY.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

eDEM-CONNECT: An Ontology-Based Chatbot for Family Caregivers of People with Dementia 12th

eDEM-CONNECT: An Ontology-Based Chatbot for Family Caregive...

引用

12th International Conference on Mobile Communication and Healthcare, MobiHealth 2023

作者： Boiting, Maurice Tschorn, Niklas Suravee, Sumaiya Yordanova, Kristina Halek, Margareta Jagoda, Franziska A. Lüdtke, Stefan Burmann, Anja Fraunhofer Institute for Software and Systems Engineering ISST Dortmund44147 Germany Institute for Data Science Universität Greifswald Greifswald17489 Germany School of Nursing Science Faculty of Health Universiät Witten/Herdecke Witten58453 Germany Institute for Visual and Analytic Computing Universät Rostock Rostock18059 Germany

ISBN: (纸本)9783031606649

Home care of people with dementia (PwD) is mainly organized and carried out by non-professional family caregivers, who struggle to interpret the needs of PwD correctly and are confronted with the challenging behavior of their relatives. Although support services for family caregivers are widespread in Germany, they are rarely used due to the fact that information is poorly organized and relatives are faced with a flood of disorganized, outdated, and confusing content. Due to the technical development of chatbot technologies (ChatGPT), chatbots gain more and more relevance. Based on the new technological possibilities, we developed an online communication and service platform with an integrated chatbot within the eDEM-CONNECT project, with the aim of making structured and easily understandable information accessible for family caregivers. This work focuses on the development of a chatbot pipeline that has broad domain knowledge through a provided ontology on the topic of agitation of PwD. This allows the chatbot to provide relevant and peer-reviewed information to family members. In our approach, a patient history is first taken based on several diagnostic questions so that relevant information can be output in a later step. For this purpose, we demonstrate that agitations in natural language can be correctly recognized by the used BERT model and that our developed chatbot is able to select further diagnostic questions based on the predictions of a Markov logic network. © ICST institute for Computer sciences, Social Informatics and Telecommunications Engineering 2024.

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

SpeedLoader: an I/O efficient scheme for heterogeneous and distributed LLM operation 24

SpeedLoader: an I/O efficient scheme for heterogeneous and d...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Yiqi Zhang Yang You Institute of Data Science National University of Singapore Singapore School of Computing National University of Singapore Singapore

ISBN: (纸本)9798331314385

With the surging growth of model parameters, foundation models pose unprecedented challenges to traditional computational infrastructures. These large models inherently require substantial accelerator memory to accommodate massive tensors during pre-training, fine-tuning, and even inference stages, making it even more challenging to deploy a model with restricted computational resources. Given this challenge, distribution and offloading the model states are two major solutions. Partitioning the required states to participating workers, and storing them in lower speed media, such as host DRAM and block devices, largely alleviate the accelerator memory pressure. However, the prohibitive costs of tensor communication render it a theoretically plausible yet practically inefficient solution. Previous efforts to improve efficiency include maximizing rematerialization and employing chunk-based tensor management to reduce host-device communication. Despite these efforts, the reported training throughput only achieves 36.54% of model FLOPs utilization (MFUs), still not comparable to full on-device training. In this work, we redesign the data flow of heterogeneous hardware and sharded model training to minimize the excessive communication overhead. Our proposed scheme significantly enhances training and inference throughput of large language models under restrictive computational resources. We confirmed a large leap in effective compute time by looking into the kernel-level runtime behavior of our trials, where the MFUs can achieve up to 51%. Compared to the state-of-the-art approach, our framework robustly achieves remarkable speedups from 3x to 30x in multiple distributed heterogeneous training setups and inference speedups of 1.5x to 2.35x without compromising arithmetic precision.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing PAN Card Security with OCR Based Information Extraction and Face Recognition for Reliable Identity Verification

Enhancing PAN Card Security with OCR Based Information Extra...

引用

data science, Agents & Artificial Intelligence (ICDSAAI), International Conference on

作者： D N Harshavardhan Reddy Malapati Sai Sravanth Reddy Panyam Sudha Dandapani Computer Science and Engineering With Data Science Sathyabama Institute of Science and Technology Chennai India Computer Science and Engineering School of Computing Sathyabama Institute of Science and Technology Chennai India

ISBN: (数字)9798331537555

ISBN: (纸本)9798331537562

Due to the potential for impersonation, the Indian PAN card is of great importance and thus, it should be encrypted with certain methods. The paper proposes a system that deals for achieving greater security of PAN card by applying a combination of OCR based information extraction and face recognition system as a means of securing identity verification. A set of OCR algorithms is first entrusted with the task to accurately read and translate the text found on the PAN card which includes the, date of birth, name and the PAN number, along with other features. The digitized and translated data is easier to store and retrieve, which makes the verification process quicker. Furthermore, a deep learning face recognition module is added in a bid to match the face of the PAN card holder with the face on the card. The proposed system is based on the architecture of the convolutional neural network (CNN) which has been shown to perform well in systems that need to compare in different amounts of light and at different angles making it reduce both the false acceptance and the false rejection rates. This article emphasizes the need for cutting-edge technologies to be implemented in identity verification systems, as identity fraud can be avoided through the use of Computerized Optical character recognition technology (OCR) combined with face recognition technology that provides advanced scanning capabilities advanced scanning capabilities that assist in safeguarding the PAN card.

关键词： Privacy Translation Face recognition Optical character recognition Information retrieval Fraud Convolutional neural networks Reliability Protection Next generation networking

来源：评论

学校读者我要写书评

暂无评论

Facial Recognition for Humanitarian Efforts: A Deep Learning based Solution for Missing Person Identification

Facial Recognition for Humanitarian Efforts: A Deep Learning...

引用

Advances in computing, Communication and Applied Informatics (ACCAI), International Conference on

作者： Dhruv Limbani Abhishek Barhate D Hemavathi Department of Data Science and Business Systems School of Computing SRM Institute of Science and Technology Kattankulathur Chennai India

ISBN: (数字)9798350389449

ISBN: (纸本)9798350389456

Our work offers a novel method for identifying missing persons that makes use of recent developments in face recognition technology. Conventional techniques depend on converting images into embeddings and then figuring out how far apart picture embeddings are from one another by calculating distances between them, making it computationally demanding, particularly while dealing with huge databases. Our process involves first extracting faces from images using OpenCV’s face extractor module, and then using VGGFace to turn those faces into embeddings. However, we employ Principal Component Analysis (PCA) to further reduce their dimensionality to 128-dimensional vectors rather than directly comparing embeddings. On the basis of these reduced embeddings, a Dense Neural Network is then trained. With this method, the traditional O(n) time for classification is greatly reduced to O(1). Our tests show that our methodology outperforms numerous other methods and reaches an excellent accuracy of 98.75%. This simplified method gives encouraging outcomes in situations involving missing persons identification in addition to increasing efficiency.

关键词： Deep learning Adaptation models Accuracy Face recognition Computational modeling Predictive models Vectors

来源：评论

学校读者我要写书评

暂无评论

Bayesian classifier based on cancer prognostic markers using accelerated failure time model with frailty effect

引用

Quality and Quantity 2025年 1-27页

作者： Vishwakarma, Gajendra K. Kumari, Pragya Bhattacharjee, Atanu Ong, Seng Huat Department of Mathematics & Computing Indian Institute of Technology Dhanbad Jharkhand Dhanbad 826004 India Division of Population Health and Genomics School of Medicine University of Dundee Dundee United Kingdom Institute of Actuarial Science and Data Analytics UCSI University Kuala Lumpur 56000 Malaysia Institute of Mathematical Sciences University of Malaya Kuala Lumpur 50603 Malaysia

This work presents a Bayesian classifier technique to categorize patients based on predictive biomarkers of time-to-event data utilizing the Accelerated Failure Time (AFT) model incorporating the frailty effect. Before classification, efficient and significant markers from a high-dimensional gene expression dataset need to be identified. Currently, it is an emerging area in oncology. A conventional three-step feature selection approach is introduced to select the most relevant markers from high-dimensional data. A threshold value for each selected marker is obtained using the proposed Bayesian classification procedure incorporating the AFT model with the frailty effect. The frailty effect is incorporated to account for unobserved heterogeneity in the expression values of subjects, allowing for a more accurate investigation of the risk of cancer-related outcomes. A simulation study is performed to validate the proposed classification approach, and the classification’s performance is evaluated using the Brier score, and the result indicates that it is relatively high. Application of the proposed classification technique is provided for two high-dimensional TCGA datasets of lung cancer patients. Our results provide evidence about the effect of classified gene expression values on the survival risk of lung cancer patients. © The Author(s), under exclusive licence to Springer Nature B.V. 2025.

关键词： Accelerated failure time model Bayesian Cancer’s biomarker Classification Survival analysis

来源：评论

学校读者我要写书评

暂无评论

Anonymous Crime Reporting using Blockchain and Smart Contract

Anonymous Crime Reporting using Blockchain and Smart Contrac...

引用

Pervasive computing and Social Networking (ICPCSN), International Conference on

作者： Prince Kumar Chaurasia Darshnilsinh Rana Rajaram V S. Srividhya Department of Networking and Communications School of Computing SRM Institute of Science and Technology Kattankulathur India Department of Data Science and Business Systems School of Computing SRM Institute of Science and Technology Kattankulathur India

ISBN: (数字)9798350386349

ISBN: (纸本)9798350386356

The traditional crime reporting systems face challenges such as inefficiencies, security vulnerabilities, and witness reluctance. Despite recent advancements like mobile applications and web platforms, challenges persist in ensuring data security, transparency, and user anonymity. These include maintaining confidentiality, data integrity, and preventing unauthorized access. To address these issues, this paper proposes leveraging blockchain technology and smart contracts. The primary objective is to enhance the security, efficiency, and anonymity of crime reporting processes. By utilizing the decentralized and immutable nature of blockchain, the system aims to ensure data integrity, transparency, and decentralization. Smart contracts will automate functions such as complaint validation and evidence verification, streamlining the reporting process. By addressing challenges in maintaining confidentiality, data integrity, and access control, the proposed solution seeks to revolutionize crime reporting, promoting trust, transparency, and accountability.

关键词： Pervasive computing Technological innovation Law enforcement Social networking (online) data integrity Smart contracts Robustness

来源：评论

学校读者我要写书评

暂无评论

Neurosymbolic AI for Mining Key Aspects of Socially Responsible Investing

Neurosymbolic AI for Mining Key Aspects of Socially Responsi...

引用

IEEE International Conference on data Mining Workshops (ICDM Workshops)

作者： Wihan Van Der Heever Ranjan Satapathy Erik Cambria Rick Siow Mong Goh College of Computing and Data Science Nanyang Technological University Singapore Institute of High Performance Computing (IHPC) Agency for Science Technology and Research (A*STAR) Singapore

ISBN: (数字)9798331530631

ISBN: (纸本)9798331530648

Environmental, Social, and Governance (ESG) factors have become critical for assessing corporate sustainability and ethical responsibility. However, the vast volume of unstructured data available across corporate reports, social media, and news sources poses a challenge for systematic ESG analysis. This paper explores the application of neurosymbolic AI, which combines neural networks’ pattern recognition capabilities with the structured reasoning of symbolic AI, to mine key aspects of ESG from large-scale, diverse data sources. By leveraging SenticNet for concept parsing and deep learning for sentiment analysis, we extract relevant ESG metrics, classify corporate practices, and identify trends. This hybrid approach enhances both the interpretability and scalability of ESG analysis, providing more accurate insights into corporate behaviors and their impact on sustainability goals. Results demonstrate that neurosymbolic AI not only improves the extraction of meaningful ESG aspects but also enables real-time monitoring, supporting data-driven decision-making for investors, regulators, and stakeholders.

关键词： Ethics Technological innovation Sentiment analysis Systematics Decision making Companies Real-time systems Stakeholders Artificial intelligence Sustainable development

来源：评论

学校读者我要写书评

暂无评论

Spatial-Temporal Graph-Based AU Relationship Learning for Facial Action Unit Detection

Spatial-Temporal Graph-Based AU Relationship Learning for Fa...

引用

2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Wang, Zihan Song, Siyang Luo, Cheng Zhou, Yuzhi Wu, Shiling Xie, Weicheng Shen, Linlin Shenzhen University Computer Vision Institute School of Computer Science & Software Engineering China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology China University of Leicester School of Computing and Mathematical Sciences United Kingdom

ISBN: (纸本)9798350302493

This paper presents our Facial Action Units (AUs) detection submission to the fifth Affective Behavior Analysis in-the-wild Competition (ABAW). Our approach consists of three main modules: (i) a pre-trained facial representation encoder which produce a strong facial representation from each input face image in the input sequence;(ii) an AU-specific feature generator that specifically learns a set of AU features from each facial representation;and (iii) a spatio-temporal graph learning module that constructs a spatio-temporal graph representation. This graph representation describes AUs contained in all frames and predicts the occurrence of each AU based on both the modeled spatial information within the corresponding face and the learned temporal dynamics among frames. The experimental results show that our approach outperformed the baseline and the spatio-temporal graph representation learning allows our model to generate the best results among all ablated systems. Our model ranks at the 4th place in the AU recognition track at the 5th ABAW Competition. Our code is publicly available at https://***/wzh125/ABAW-5. © 2023 IEEE.

关键词： Graphic methods

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：