检索结果-内蒙古大学图书馆

AutoQNN: An End-to-End Framework for Automatically Quantizing Neural Networks

Journal of computer science & technology 2024年第2期39卷 401-420页

作者：龚成卢冶代素蓉邓倩杜承昆李涛 College of Software Nankai UniversityTianjin 300350China College of Computer Science Nankai UniversityTianjin 300350China State Key Laboratory of Processors Institute of Computing TechnologyChinese Academy of Sciences Beijing 100190China

Exploring the expected quantizing scheme with suitable mixed-precision policy is the key to compress deep neural networks(DNNs)in high efficiency and *** exploration implies heavy workloads for domain experts,and an automatic compression method is ***,the huge search space of the automatic method introduces plenty of computing budgets that make the automatic process challenging to be applied in real *** this paper,we propose an end-to-end framework named AutoQNN,for automatically quantizing different layers utilizing different schemes and bitwidths without any human *** can seek desirable quantizing schemes and mixed-precision policies for mainstream DNN models efficiently by involving three techniques:quantizing scheme search(QSS),quantizing precision learning(QPL),and quantized architecture generation(QAG).QSS introduces five quantizing schemes and defines three new schemes as a candidate set for scheme search,and then uses the Differentiable Neural Architecture Search(DNAS)algorithm to seek the layer-or model-desired scheme from the *** is the first method to learn mixed-precision policies by reparameterizing the bitwidths of quantizing schemes,to the best of our *** optimizes both classification loss and precision loss of DNNs efficiently and obtains the relatively optimal mixed-precision model within limited model size and memory *** is designed to convert arbitrary architectures into corresponding quantized ones without manual intervention,to facilitate end-to-end neural network *** have implemented AutoQNN and integrated it into *** experiments demonstrate that AutoQNN can consistently outperform state-of-the-art *** 2-bit weight and activation of AlexNet and ResNet18,AutoQNN can achieve the accuracy results of 59.75%and 68.86%,respectively,and obtain accuracy improvements by up to 1.65%and 1.74%,respectively,compared with state-of-the-art ***,c

关键词： automatic quantization mixed precision quantizing scheme search quantizing precision learning quan-tized architecture generation

来源：评论

学校读者我要写书评

暂无评论

Short-Term Power Load Forecasting Method Based on GRU-Transformer Combined Neural Network Model

引用

Journal of computing and Information technology 2024年第1期32卷 1-14页

作者： Mao, Weiwei Yu, Suping Chen, Wenqing College of Computer and Information Engineering Luoyang Institute of Science and Technology Luoyang China School of Electrical Engineering and Automation Luoyang Institute of Technology Luoyang China

Load Forecast (LF) is an important task in the planning, control and application of public power systems. Accurate Short Term Load Forecast (STLF) is the premise of safe and economical operation of a power system. In the research of short-term power load forecasting, machine learning and deep learning are the most popular methods at present, but there still exists a problem that the single and simple structure of power load forecasting model leads to low accuracy of load forecasting. In order to improve the accuracy of STLF, a Gated Cycle Unit (GRU)-Transformer combined neural network model is proposed. Transformer encoder structure is used as feature extractor to mine the complex mapping relationships between the input features and load. The advantage of self-attention mechanism is used to solve the problem of information loss of long sequences in short-term power load forecasting. At the same time, the multivariate time series model of GRU is used for model training. The experimental results on the power load data set of a certain region in southwest China and Panama City show that the proposed combined model prediction method has higher accuracy than those proposed in other literatures, which further proves its feasibility and superiority. © 2024, University of Zagreb Faculty of Electrical Engineering and computing. All rights reserved.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Hash-Indexing Block-Based Deduplication Algorithm for Reducing Storage in the Cloud

引用

computer Systems science & Engineering 2023年第7期46卷 27-42页

作者： D.Viji S.Revathy Department of Computer Science and Engineering Sathyabama Institute of Science and TechnologyChennaiIndia

Cloud storage is essential for managing user data to store and retrieve from the distributed data *** storage service is distributed as pay a service for accessing the size to collect the *** to the massive amount of data stored in the data centre containing similar information and file structures remaining in multi-copy,duplication leads to increase storage *** potential deduplication system doesn’t make efficient data reduction because of inaccuracy in finding similar data *** creates a complex nature to increase the storage consumption under *** resolve this problem,this paper proposes an efficient storage reduction called Hash-Indexing Block-based Deduplication(HIBD)based on Segmented Bind Linkage(SBL)Methods for reducing storage in a cloud ***,preprocessing is done using the sparse augmentation ***,the preprocessed files are segmented into blocks to make *** block of the contents is compared with other files through Semantic Content Source Deduplication(SCSD),which identifies the similar content presence between the *** on the content presence count,the Distance Vector Weightage Correlation(DVWC)estimates the document similarity weight,and related files are grouped into a ***,the segmented bind linkage compares the document to find duplicate content in the cluster using similarity weight based on the coefficient match *** implementation helps identify the data redundancy efficiently and reduces the service cost in distributed cloud storage.

关键词： Cloud computing deduplication hash indexing relational content analysis document clustering cloud storage record linkage

来源：评论

学校读者我要写书评

暂无评论

An intrusion detection system using ranked feature bagging

引用

International Journal of Information technology (Singapore) 2024年第2期16卷 1213-1219页

作者： Azhagiri, M. Rajesh, A. Karthik, S. Raja, K. Computer Science and Engineering SRM Institute of Science and Technology Ramapuram Chennai India Department of Computer Science and Engineering Jain University Bangalore India Department of IT College of Computing and Informatics Saudi Electronic University Riyadh Saudi Arabia Department of Computer Science and Engineering SRM Institute of Science and Technology Ramapuram Chennai India

An Intrusion Detection System monitors the network for any malicious attacks. It is an ideal tool for protecting extensive business networks from any kind of attack. In this paper, an Intrusion Detection System using Bagging and ranked features has been proposed. The sample datasets for training the ensemble of classifiers in the bagging approach are obtained by generating subsets of the features ranked according to their importance in the classification process. OneR algorithm has been used to rank the features. The decision tree has been used as the base classifier in the bagging approach. The proposed system was able to predict attacks with an accuracy of 99.71% with a minimal number of features and outperformed other similar approaches discussed in the literature. © The Author(s), under exclusive licence to Bharati Vidyapeeth's institute of computer Applications and Management 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s);author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

关键词： Bagging CIC-IDS 2017 dataset Decision tree Ensemble Feature selection Intrusion detection system OneR feature selection

来源：评论

学校读者我要写书评

暂无评论

Contemporary Approaches to Analyze Non-stationary Time-series: Some Solutions and Challenges

引用

Recent Advances in computer science and Communications 2023年第2期16卷 61-80页

作者： Dixit, Ankit Jain, Shikha Department of Computer Science & Engineering and Information Technology Jaypee Institute of Information Technology Noida India

Enhancement of technology yields more complex time-dependent outcomes for better understanding and analysis. These outcomes generate more complex, unstable, and high-dimensional data from non-stationary environments. Hence, more challenges are arising day by day to fulfill the increasing demand for future estimation. Thus, in this paper, an extensive study has been presented to comprehend the statistical complexity and randomness of Non-Stationary Time Series (NS-TS) data at the atomic level. This survey briefly explains the basic principles and terms related to Non-Stationary Time Series (NS-TS). After understanding the fundamentals of NS-TS, this survey categorized non-stationary time series into groups and subgroups based on a change in statistical behavior. It also provides a comprehensive discussion on contemporary approaches proposed by researchers in each category of non-stationarity. These algorithms include clustering, classification, and regression techniques to deal with different types of domains. Every category of non-stationarity consists of a separate table to draw some advantages and disadvantages of existing approaches. At the end of each non-stationarity type, a short discussion and critical analysis have been done. In the conclusion section, it is observed that this research sphere still has many open challenges that need to be addressed and demand more exploration. Furthermore, it discusses the possible solution of improvisation in future research. © 2023 Bentham science Publishers.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

WeaponVision AI: a software for strengthening surveillance through deep learning in real-time automated weapon detection

引用

International Journal of Information technology (Singapore) 2025年第3期17卷 1717-1727页

作者： Yadav, Pavinder Gupta, Nidhi Sharma, Pawan Kumar Department of Computer Applications National Institute of Technology Haryana Kurukshetra 136119 India Mathematics and Scientific Computing National Institute of Technology HP Hamirpur India

During situations involving dangerous activities, such as armed robbery in public areas, surveillance systems often exhibit delays or inefficiencies in their prompt responding. To obviate the necessity for human involvement, there is a requirement for technology capable of autonomously detecting harmful objects, like firearms in surveillance footages. This article presents WeaponVision AI, an advanced software system that has the ability to accurately identify weapons in live feeds, recorded videos, and images. Moreover, this software has the capability to detect guns even under weak lighting circumstances. The deep learning architecture based on modified YOLOv7 was trained on a vast dataset assortment of 79,558 images of weapons, in developing the WeaponVision AI. The model exhibited satisfactory results following the training phase, achieving significant outcome metrics: a precision rate of 91.75% and a mean average precision of 92.15%. The efficacy of WeaponVision AI is showcased through its ability to accurately identify weapons across diverse environmental and visual conditions. © Bharati Vidyapeeth's institute of computer Applications and Management 2025.

关键词： computer vision Deep learning Machine learning Security and surveillance Surveillance system Weapon detection

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approach for Identifying Red Lesions in Retinal Fundus Imagery 6th

Deep Learning Approach for Identifying Red Lesions in Retina...

引用

6th International conference on communication and computational technologies, ICCCT 2024

作者： Divyashrree, R. Dhawan, Samarth Saranya, P. Department of Computing Technologies School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai603203 India

ISBN: (纸本)9789819774258

One dangerous side effect of diabetes that affects the eyes is called diabetic retinopathy. It happens as a result of alterations in the retina’s blood vessels, which can cause harm and even blindness. The development of red lesions in the retina is one typical sign of diabetic retinopathy. These red lesions may be a sign of microaneurysms or hemorrhages, which are important characteristics to identify and track as diabetic retinopathy progresses. This process becomes even more demanding in the early stages of the disease where symptoms are less pronounced in the images. Leveraging machine-based learning for medical image analysis, especially deep learning algorithms, has demonstrated its capability to enhance the early diagnosis of diabetic retinopathy, providing a more efficient and effective alternative to traditional manual assessments. The proposed work aims to improve the efficiency of early detection of diabetic retinopathy by utilizing Deep Learning (16-layer CNN) to determine a person’s likelihood of having diabetic retinopathy using a 02-Class model that compiles a variety of datasets, specifically IDRiD dataset. The entire dataset contains 18,590 fundus photographs, in which 3662 are allocated for training, 1928 for validation, and 13,000 for testing. All datasets had a similar distribution of output classes, which is a fundamental property of this type of data. Using the IDRiD dataset, the suggested Red Lesion detection system’s specificity and sensitivity were found to be 98% and 89%, respectively, with an accuracy of 95.32%. Using the MESSIDOR dataset, the Diabetic Retinopathy Severity Classification System’s specificity, sensitivity, and accuracy were found to be 93.8, 92.3, and 94%, respectively. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Ophthalmology

来源：评论

学校读者我要写书评

暂无评论

Spider monkey ınspired optimization algorithm based clustering for face recognition

引用

International Journal of Information technology (Singapore) 2024年 1-13页

作者： Gopal, Veni devi Devi, G. Shree Geetha, Angelina Department of Computer Science and Engineering B S Abdur Rahman Institute of Science & amp Technology Chennai India Department of Computer Science and Engineering Hindustan Institute of Science and Technology Chennai India Dept of Computer Applications BSA Crescent Institute of Science and Technology Chennai India

Face recognition is a fast-growing technology that is widely used in forensics such as criminal identification, secure access, and prison *** contrasts from other classification issues in that there are normally a more number of classes in face recognition. Thus, the validation accuracy of the conventional subsystem face recognition is unsuitable. In order to overcome the problems that are likely to be encountered, the proposed method utilizes the Spider Monkey inspired Optimization (SMO) algorithm for face clustering and recognition. At first feature reduction is done by Advance Principal Component Analysis (APCA) which is a hybridization of PCA and linear discriminant analysis (LDA). After reducing the features, the selected features are grouped and faces are recognized by means of SMO algorithm. From the food search behavior of spider monkeys, SMO algorithm is developed. The performance of the proposed method is evaluated by means of sensitivity, specificity, and accuracy. Here the proposed method accomplishes 98.66% accuracy. Using the MATLAB platform, the recommended technique is implemented. © Bharati Vidyapeeth's institute of computer Applications and Management 2024.

关键词： Accuracy Linear discriminant analysis Principal component analysis Sensitivity Specificity Spider monkey optimization

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Models Based on Weakly Supervised Learning and Clustering Visualization for Disease Diagnosis

引用

computers, Materials & Continua 2023年第9期76卷 2649-2665页

作者： Jingyao Liu Qinghe Feng Jiashi Zhao Yu Miao Wei He Weili Shi Zhengang Jiang School of Computer and Information Engineering Chuzhou UniversityChuzhou239000China School of Computer Science and Technology Changchun University of Science and TechnologyChangchun130022China Zhongshan Institute of Changchun University of Science and Technology ZhongshanChina School of Intelligent Engineering Henan Institute of TechnologyXinxiang453003China

The coronavirus disease 2019(COVID-19)has severely disrupted both human life and the health care *** diagnosis and treatment have become increasingly important;however,the distribution and size of lesions vary widely among individuals,making it challenging to accurately diagnose the *** study proposed a deep-learning disease diagnosismodel based onweakly supervised learning and clustering visualization(W_CVNet)that fused classification with ***,the data were *** optimizable weakly supervised segmentation preprocessing method(O-WSSPM)was used to remove redundant data and solve the category imbalance ***,a deep-learning fusion method was used for feature extraction and classification recognition.A dual asymmetric complementary bilinear feature extraction method(D-CBM)was used to fully extract complementary features,which solved the problem of insufficient feature extraction by a single deep learning ***,an unsupervised learning method based on Fuzzy C-Means(FCM)clustering was used to segment and visualize COVID-19 lesions enabling physicians to accurately assess lesion distribution and disease *** this study,5-fold cross-validation methods were used,and the results showed that the network had an average classification accuracy of 85.8%,outperforming six recent advanced classification models.W_CVNet can effectively help physicians with automated aid in diagnosis to determine if the disease is present and,in the case of COVID-19 patients,to further predict the area of the lesion.

关键词： Classification COVID-19 deep learning segmentation unsupervised learning weakly supervised

来源：评论

学校读者我要写书评

暂无评论

Influence of different form of multimedia on motivation and topic perception in augmented reality learning experiences

引用

Multimedia Tools and Applications 2025年 1-28页

作者： Yap, Shook Chin Rahmat, Rahmita Wirza Ali, Siti Khadijah School of Computing and Data Science Xiamen University Malaysia Selangor Sepang Malaysia Faculty of Computer Science and Information Technology Universiti Putra Malaysia Selangor Serdang Malaysia

It has been widely proven that Augmented Reality (AR) brings numerous benefits in learning experiences, including enhancing learning outcomes and motivation. However, not many studies investigate how different forms of multimedia help in motivating students in AR learning experiences. Instead of examining whether AR can be used in education to motivate students, our research focused on investigating how and when AR learning experiences influence motivation when integrated with multimedia learning. A between-subjects 2 × 3 factorial design was employed in this study. The first factor was the type of AR (static-AR or animated-AR), and the second factor was the modality of the description (concurrent narration-text, text only, or narration only). Six experimental conditions were designed, each equipped with different versions of AR learning applications with various combinations of multimedia formats. One hundred and forty-four computer science and information technology undergraduate students (70 males and 74 females) were randomly assigned to one of the six experimental conditions. Data collection was conducted with questionnaires, and data were analyzed using univariate analysis of variance (ANOVA). Results show that attention and satisfaction subscales of the ARCS motivational model can be enhanced with 3D animation implemented in AR. Descriptions delivered with concurrent narration and written text in AR are effective in enhancing motivation as well. Besides, participants’ perceived understanding and interest in learning the topic increased after AR learning experiences. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Contrastive Learning

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：