检索结果-内蒙古大学图书馆

International conference of Electronics, Communication and Aerospace Technology (ICECA)

作者： Sripriya Arunachalam Shubhangi N. Ghate N. Kopperundevi M.S Balamurugan D. Kanagajothi Srimathi S Computer Science and Application SRM Institute of Science and Technology Vadapalani Campus College of Science and Humanities Electronics and Telecommunications Engg. Ramrao Adik Institute of Technology D. YiPatil University Navi-Mumbai Maharashtra India School of Computer Science and Engineering Vellore Institute of Technology Vellore Tamil Nadu Department of ECE Panimalar Engineering College Department of Mathematics Vel Tech Rangarajan Dr Shangunthala R&D Institute of Science and Technology Department of Bio Technology Saveetha School of Engineering SIMATS Chennai Tamilnadu India

One of the most frequent and debilitating mental illnesses is major depression. Video-based identification of depression using facial expressions has been proposed by a number of deep learning architectures. These structures are frequently modeled for regression with Euclidean loss to foretell the degree of depression. The ordinal correlation between depressive symptoms and exposure to facial depictions is not explored, and the models' resilience to noisy and ambiguous labeling is constrained as a result. Famous deep architectures have shown guarantee in facial depression acknowledgment, but they still lack proper discriminative power because of problems with a lack of labeled depression information for deep transfer learning, large variations in facial expression between people who have identical symptom severity, and subtle variations in facial expression between varying depression severities. This study models the challenge of facial depression identification as one of the learnings to distribute labels, and this study offers a deep combined label distribution and metric learning approach to this problem. Using distribution learning, this research study provides a deep learning architecture for predicting depressive symptoms. The proposed deep learning on spatiotemporal feature improves the process of depression prediction because the two learning modules work together to improve the feature's representation ability and discriminatory strength. The results of empirical evaluation of the proposed method on publicly available datasets prove the system efficacy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Predicting the Risk of Illegal Activity and Evaluating Law Enforcement Using WideDeep SGRU Model

A Novel Approach to Predicting the Risk of Illegal Activity ...

引用

Electronics, Computing, Communication and Control Technology (ICECCC), International Conference on

作者： Prativa Mishra Anusha P Arvind Kumar Subharun Pal Jyoti Prasad Patra Amit Chauhan Computer Science and Application Reva University Bangalore Karnataka India Department of ECE R.M.K. Engineering College Kavaraipettai Tamilnadu India SCA Manav Rachna International Institute of Research and Studies Faridabad Haryana Computer Science and Engineering Indian Institute of Technology Jammu Jagti Jammu & Kashmir (UT) India EE And EEE Krupajal Engineering College-KEC Puri Odisha India Department of Life sciences School of Sciences CHRIST (Deemed to be University) Bengaluru Karnataka India

ISBN: (数字)9798350371802

ISBN: (纸本)9798350371819

The main reaction to the illicit extraction of natural resources in protected areas around the world is law enforcement patrols headed by rangers. On the other hand, research on patrols' efficacy in reducing criminal behavior is lacking. Similarly, tactics to enhance the effectiveness of patrol organization and monitoring have received very little attention. Sequencing is crucial for model training, feature selection, and preprocessing. Preprocessing steps include cleaning, discretizing, duplicating, and normalizing data. Feature selection makes use of genetic algorithms, which are basically search algorithms with an evolutionary bent that factor in natural selection and genetics. Training stacked GRU models necessitates meticulous feature management. Even the most cutting-edge algorithms, GRU and BiGRU, are no match for the suggested technique. An astounding 97.24% accuracy grade was disclosed by the results, showcasing exceptional growth.

关键词： Training Sequential analysis Accuracy Law enforcement Computational modeling Feature extraction Cleaning

来源：评论

学校读者我要写书评

暂无评论

Effects of chelating agents on the sol-gel synthesis of nano-zirconia:Comparison of the Pechini and sugar-based methods

引用

International Journal of Minerals,Metallurgy and Materials 2020年第5期27卷 693-702页

作者： Faramarz Kazemi Farzin Arianpour Mahdiar Taheri Ali Saberi Hamid Reza Rezaie Department of Mining and Metallurgy Amirkabir University of TechnologyTehranIran Research and Application Center Kastamonu UniversityKastamonuTurkey College of Engineering and Computer Science The Australian National UniversityCanberraAustralia Materials Processing Faculty of Engineering ScienceUniversity of BayreuthBayreuthGermany School of Metallurgy and Materials Engineering Iran University of Science and TechnologyTehranIran

This study focused on the comparison of the Pechini and sugar-based combustion synthesis methods to produce nano-zirconia. Zirconium hydroxide was utilized as metal precursor and citric acid, sucrose, and fructose were used as chelating agents, followed by calcination at 500, 600, and 700°C in air, respectively. Characterization was conducted by thermal analysis, specific surface area measurement, Fourier transform infrared spectroscopy, X-ray diffraction, and scanning and transmission electron microscopy. When sucrose and citric acid were used as chelating agents during synthesis, mixtures of monoclinic and tetragonal phases were formed after calcination at 600 and 700°C. In the fructose samples, the tetragonal structure was the unique characterized phase. The tetragonal parameters in the fructose samples were determined using the diffraction data and the lattice parameter ratio was proven to increase with the temperature increase. Compared with the citrate and sucrose samples, the largest specific surface area(27 m2·g-1) and smallest particle size(39.1 nm) were obtained for the fructose sample after calcination at 700°C. The study revealed the formation of single-phase stabilized tetragonal zirconia using fructose as chelating agent after calcination at 500°C, and the presence and formation mechanism of stabilized tetragonal phase were also discussed on the basis of the X-ray and electron diffraction studies.

关键词： zirconia nanopowder sol-gel sugar Pechini chelation

来源：评论

学校读者我要写书评

暂无评论

Modernizing Smart Homes into Cognitive Residences

Modernizing Smart Homes into Cognitive Residences

引用

Communication, Computing and Energy Efficient Technologies (I3CEET), International Conference on

作者： Gunjan Sharma K. Ranjith Singh R.Udayakumar Haider Alabdeli Kanav Krishan Azizbek Khurramov Institute of Business Management GLA University Mathura India Department of Computer Science Karpagam Academy of Higher Education Coimbatore India CS & IT Kalinga University India Department of Computers Techniques Engineering College of Technical Engineering The Islamic University Najaf Iraq Department of Computers Techniques Engineering College of Technical Engineering The Islamic University of Al Diwaniyah Al Diwaniyah Iraq Department of Computer Application Chandigarh Engineering College Chandigarh Group of Colleges Mohali Punjab India International School of Finance and Technology Tashkent Uzbekistan

ISBN: (数字)9798331541583

ISBN: (纸本)9798331541590

The growing popularity of connected homes can be attributed to advances in cloud computing and machine learning. The latest technology allows devices such as smart speakers and thermostats to learn from your feedback and adapt to your personal preferences. However, these smart gadgets can also influence people's behavior. Investigate the potential impact of smart homes on behavior using a human model with hierarchical reinforcement learning. This model can be used for a variety of tasks, such as controlling humidity and temperature within a smart home under Q-learning management. We investigate possible behavioural changes and explore prospects for mutual adaptation between smart home and human models. Our human models are trained on these tasks using hierarchical reinforcement learning, which allows them to learn when to move on to new activities and how long to focus on existing activities. We then integrate human modelling with digital house models in real-world scenarios and conduct in-depth studies involving different situations in which one or more human models interact with a smart home. Our results suggest that installing a connected home may cause human models to exhibit unexpected behavior, such as frequent changes in activity or longer times to change temperature settings. What is interesting is that when two human samples are combined, some exhibit typical behavior while others reproduce the surprising results of single-subject experiments.

关键词： Adaptation models Q-learning Computational modeling Smart homes Humidity Switches Behavioral sciences Temperature control Telecommunication computing Thermostats

来源：评论

学校读者我要写书评

暂无评论

Quantum cryptography technique: A way to improve security challenges in mobile cloud computing (MCC)

引用

Materials Today: Proceedings 2022年 51卷 508-514页

作者： Shafiqul Abidin Amit Swami Edwin Ramirez-Asís Joseph Alvarado-Tolentino Rajesh Kumar Maurya Naziya Hussain Department of Computer Science Aligarh Muslim University Aligarh Uttar Pradesh India Emirates College of Technology United Arab Emirates Department of Administration and Tourism Santiago Antúnez de Mayolo National University USA Department of Systems and Informatics Engineering Santiago Antúnez de Mayolo National University USA Master of Computer Application ABES Engineering College Ghaziabad India Department of Computer Science and Engineering School of Computers IPS Academy Indore India

Quantum cryptography concentrates on the solution of cryptography that is imperishable due to the reason of fortification of secrecy which is applied to the public key distribution of quantum. It is a very prominent technology in which 2 beings can securely communicate along with the sights belongings to quantum physics. However, on basis of classical level cryptography, the used encodes were bits for data. As quantum utilizes the photons or particles polarize ones for encoding the quantized property. This is presented in qubits as a unit. Transmissions depend directly on the inalienable mechanic's law of quantum for security. This paper includes detailed insight into the three most used and appreciated quantum cryptography applications that are providing its domain-wide service in the field of mobile cloud computing. These services are (it) DARPA Network, (ii) IPSEC implementation, and (iii) the twisted light HD implementation along with quantum elements, key distribution, and protocols.

关键词： Quantum cryptography DARPA IPSEC Twisted light Mobile cloud computing

来源：评论

学校读者我要写书评

暂无评论

Pre-training Everywhere: Parameter-Efficient Fine-Tuning for Medical Image Analysis via Target Parameter Pre-training

arXiv

引用

arXiv 2024年

作者： Lei, Xingliang Ye, Yiwen Chen, Ziyang Shu, Minglei Xia, Yong National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Northwestern Polytechnical University China China Ningbo Institute of Northwestern Polytechnical University Ningbo315048 China Research & Development Institute of Northwestern Polytechnical University in Shenzhen Shenzhen518057 China

Parameter-efficient fine-tuning (PEFT) techniques have emerged to address issues of overfitting and high computational costs associated with fully fine-tuning in the paradigm of self-supervised learning. Mainstream methods based on PEFT involve adding a few trainable parameters while keeping the pre-trained parameters of the backbone fixed. These methods achieve comparative, and often superior, performance to fully fine-tuning, demonstrating the powerful representation ability of the pre-trained backbone. Despite its success, these methods typically ignore the initialization of the new parameters, often relying solely on random initialization. We argue that if pre-training is significantly beneficial, it should be applied to all parameters requiring representational capacity. Motivated by this insight, we propose a simple yet effective fine-tuning framework based on Target Parameter Pre-training (TPP). The target parameters refer to the new parameters introduced during fine-tuning. TPP includes an additional stage before PEFT to pre-train these target parameters. During this stage, the pre-trained backbone parameters are frozen, and only the target parameters are trainable. A defined pre-text task is used to encourage the target parameters to learn specific representations of downstream data. When PEFT is subsequently employed, the pre-trained target parameters are loaded to enhance fine-tuning efficiency. The proposed TPP framework is versatile, allowing for the integration of various pretext tasks for pre-training and supporting different PEFT methods as backbones. We evaluated the fine-tining performance of our method using five public datasets, including three modalities and two task types. The results demonstrate that the proposed TPP can be easily integrated into existing PEFT methods, significantly improving performance. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

One for All: One-stage Referring Expression Comprehension with Dynamic Reasoning

arXiv

引用

arXiv 2022年

作者： Zhang, Zhipeng Wei, Zhimin Huang, Zhongzhen Niu, Rui Wang, Peng School of Computer Science Northwestern Polytechnical University Xi’an China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology China

Referring Expression Comprehension (REC) is one of the most important tasks in visual reasoning that requires a model to detect the target object referred by a natural language expression. Among the proposed pipelines, the one-stage Referring Expression Comprehension (OSREC) has become the dominant trend since it merges the region proposal and selection stages. Many state-of-the-art OSREC models adopt a multi-hop reasoning strategy because a sequence of objects is frequently mentioned in a single expression which needs multi-hop reasoning to analyze the semantic relation. However, one unsolved issue of these models is that the number of reasoning steps needs to be pre-defined and fixed before inference, ignoring the varying complexity of expressions. In this paper, we propose a Dynamic Multi-step Reasoning Network, which allows the reasoning steps to be dynamically adjusted based on the reasoning state and expression complexity. Specifically, we adopt a Transformer module to memorize & process the reasoning state and a Reinforcement Learning strategy to dynamically infer the reasoning steps. The work achieves the state-of-the-art performance or significant improvements on several REC datasets, ranging from RefCOCO (+, g) with short expressions, to Ref-Reasoning, a dataset with long and complex compositional expressions. © 2022, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Aided Intelligent Framework for Condition Monitoring of Electrical Machinery

A Deep Learning Aided Intelligent Framework for Condition Mo...

引用

Devices for Integrated Circuit (DevIC)

作者： Biswarup Ganguly Radha Kumari Ray Arunava Chatterjee Subho Paul Electrical Engineering Department Meghnad Saha Institute of Technology Kolkata India Application Development Associate Accenture Kolkata India Electrical Engineering Department Raghunathpur Govt. Polytechnic Purulia India School of Electrical Engineering and Computer Science Washington State University Pullman WA USA

Single-phase induction motors (SPIM) are fairly efficiently employed in household devices and possess wide applications in the industrial system. Typically, current signature-analysis (CSA) based methods have been used for fault diagnosis, but not always been an efficient solution, especially for fractional kW SPIMs that may be replaced entirely iffaulty. This paper presents a deep learning (DL) based fault detection scheme of a SPIM which is in-service in household devices. The fault detection is done employing a wavelet based deep dense convolutional neural network (W-D2CNN), where several wavelets are used as convolutional filters for convolution operation. The convoluted feature maps are further passed through a proposed nonlinearity, i.e., SLReLU, for richer feature extraction. The uniqueness of the proposed method lies in the feature learning ability, where upper-and lower-layer features of the dense block fuse via concatenation operation. Also, the SL-ReLU has been designed to combine logarithmic function, softsign, and ReLU operation. The designed SL-ReLU minimizes the probability of convoluted output and gradient explosion, due to which training has been improved along with feature learning ability. Comprehensive results reveal the efficacy of the proposed work in terms of performance metrics with state-of-the-art DL frameworks using the same dataset. The proposed work can also be extendable to multiphase-induction machines or any other electrical device working in any environment for fault identification.

关键词： Representation learning Fault diagnosis Training Deep learning Fault detection Feature extraction Explosions

来源：评论

学校读者我要写书评

暂无评论

Vision-Language Consistency Guided Multi-modal Prompt Learning for Blind AI Generated Image Quality Assessment

arXiv

引用

arXiv 2024年

作者： Fu, Jun Zhou, Wei Jiang, Qiuping Liu, Hantao Zhai, Guangtao The CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China The School of Computer Science and Informatics Cardiff University CardiffCF24 4AG United Kingdom The School of Information Science and Engineering Ningbo University Ningbo315211 China The Institute of Image Communication and Network Engineering Shanghai Jiao Tong University Shanghai200240 China

Recently, textual prompt tuning has shown inspirational performance in adapting Contrastive Language-Image Pre-training (CLIP) models to natural image quality assessment. However, such uni-modal prompt learning method only tunes the language branch of CLIP models. This is not enough for adapting CLIP models to AI generated image quality assessment (AGIQA) since AGIs visually differ from natural images. In addition, the consistency between AGIs and user input text prompts, which correlates with the perceptual quality of AGIs, is not investigated to guide AGIQA. In this letter, we propose vision-language consistency guided multi-modal prompt learning for blind AGIQA, dubbed CLIP-AGIQA. Specifically, we introduce learnable textual and visual prompts in language and vision branches of CLIP models, respectively. Moreover, we design a text-to-image alignment quality prediction task, whose learned vision-language consistency knowledge is used to guide the optimization of the above multi-modal prompts. Experimental results on two public AGIQA datasets demonstrate that the proposed method outperforms state-of-the-art quality assessment models. The source code is available at https://***/JunFu1995/CLIP-AGIQA. © 2024, CC BY-NC-ND.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Study of Natural Language Processing for Sentiment Analysis

Study of Natural Language Processing for Sentiment Analysis

引用

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

作者： Kavitha C.R. Rajarajan S J R. Jothilakshmi Kamal Alaskar Mohammad Ishrat V. Chithra Devi Department of Computer Science Engineering Amrita School of Computing Bengaluru Amrita Vishwa Vidyapeetham India Department of Mechanical Engineering S.A. Engineering College (Autonomous) Thiruverkadu Tamil Nadu India Department of Information Technology R.M.D. Engineering College Kavaraipettai Tamil Nadu India Department of Computer Application BharatiVidyapeeth (Deemed to be University) Institute of Management Kolhapur Maharashtra India Department of Computer Science and Engineering KoneruLakshmaiah Education Foundation Vaddeswaram Guntur Andhra Pradesh India Department of English PSNA College of Engineering and Technology (Autonomous) Dindigul Tamil Nadu India

The study of opinions, sentiments, assessments, perspectives, and feelings as they are expressed in written language is the focus of sentiment analysis and evaluation mining. It is commonly the subject of text mining, information mining, and web mining because it is one of the most active study fields in natural language processing. Sentiment analysis has been applied in many different contexts, including the study of how events affect unofficial groups and the exploration of presumptions about goods and services. The importance of sentiment analysis is increasing at the same time as virtual entertainment such as surveys, group chats, web blogs, micro sites, Twitter, and interpersonal organizations. In this work, we provide the results of our preliminary studies on twitter sentiment analysis. This study seeks to distinguish subjects from sentiment in tweets. It recognizes the sentiment that refers to the particular problem using Natural Language Processing techniques. Our analysis consists of three main processes, subjective grouping, semantic affiliation, and extreme order, to characterize sentiment. By describing the syntactic relationship between sentiment dictionaries and subject, the study makes use of sentiment vocabularies. Exploratory results demonstrate that the proposed framework performs better than existing message sentiment analysis tools because tweets don't have the same architecture as traditional messages.

关键词：

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：