检索结果-内蒙古大学图书馆

The Journal of Machine Learning Research 2024年第1期25卷 2506-2556页

作者： Wolfgang Roth Günther Schindler Bernhard Klein Robert Peharz Sebastian Tschiatschek Holger Fröning Franz Pernkopf Zoubin Ghahramani Graz University of Technology Austria Laboratory of Signal Processing and Speech Communication Heidelberg University Germany Institute of Computer Engineering Graz University of Technology Austria Institute of Theoretical Computer Science University of Vienna Austria Faculty of Computer Science University of Cambridge UK

While machine learning is traditionally a resource intensive task, embedded systems, autonomous navigation, and the vision of the Internet of Things fuel the interest in resource-efficient approaches. These approaches aim for a carefully chosen trade-off between performance and resource consumption in terms of computation and energy. The development of such approaches is among the major challenges in current machine learning research and key to ensure a smooth transition of machine learning technology from a scientific environment with virtually unlimited computing resources into everyday's applications. In this article, we provide an overview of the current state of the art of machine learning techniques facilitating these real-world requirements. In particular, we focus on resource-efficient inference based on deep neural networks (DNNs), the predominant machine learning models of the past decade. We give a comprehensive overview of the vast literature that can be mainly split into three non-mutually exclusive categories: (i) quantized neural networks, (ii) network pruning, and (iii) structural efficiency. These techniques can be applied during training or as post-processing, and they are widely used to reduce the computational demands in terms of memory footprint, inference speed, and energy efficiency. We also briey discuss different concepts of embedded hardware for DNNs and their compatibility with machine learning techniques as well as potential for energy and latency reduction. We substantiate our discussion with experiments on well-known benchmark data sets using compression techniques (quantization, pruning) for a set of resource-constrained embedded systems, such as CPUs, GPUs and FPGAs. The obtained results highlight the difficulty of finding good trade-offs between resource efficiency and prediction quality.

关键词： resource-efficient machine learning inference deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Web Application for Real-Time Apple Leaf Disease Detection and Classification

Deep Learning-Based Web Application for Real-Time Apple Leaf...

引用

Advancements in Smart, Secure and Intelligent Computing (ASSIC), International Conference on

作者： Satish Kumar Rakesh Kumar Meenu Gupta Ahmed J. Obaid Department of Computer Science and Engineering Chandigarh University Punjab India Faculty of Computer Science and Mathematics University of Kufa Najaf Iraq Department of Computer Technical Engineering Technical Engineering College Al-Ayen University Thi-Qar Iraq

Agriculture is the most significant industry in the economy of India. Various kinds of diseases affect the leaves of plants and influence the productivity of crops. Apple farmers are also constantly facing challenges in boosting their yield and protecting apple trees from diseases. The prevalence of diseases and pests significantly hampers apple production, leading to substantial financial losses for the industry each year. Farmers may not have expertise in leaf disease prediction. Detecting Apple Leaf Diseases (ALD) swiftly and accurately is crucial for effectively handling and curbing these issues within orchards. Specifically, advancements in computer vision methods utilizing Deep Learning (DL) have opened up avenues for identifying and understanding these diseases at an early stage directly on the leaves. Web application based on the DL model is proposed to address this issue, which can predict Healthy and Alternaria, Leaf Spot, Marssonina Blotch, and Powdery mildew disease of the affected leaf.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Applying Machine Learning Techniques to Increase Real-Time Data Analysis Accuracy 1

Applying Machine Learning Techniques to Increase Real-Time D...

引用

1st IEEE International Conference on Emerging Research in Computational science, ICERCS 2023

作者： Naval, Preeti Mahesh, T.R. Kumar, Ajay Dhingra, Lovish Janakiraman, V. Banchhor, Chitrakant O. Maharishi University of Information Technology Maharishi School of Engineering and Technology Uttar Pradesh India Faculty of Engineering and Technology Department of Computer Science Engineering Karnataka Bangalore India Vivekananda Global University Department of Computer Science & Engineering Jaipur India Chitkara University Institute of Engineering and Technology Chitkara University Centre of Interdisciplinary Research in Business and Technology Punjab India Prince Shri Venkateshwara Padmavathy Engineering College Department of Mechanical Engineering Chennai127 India Vishwakarma Institute of Information Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350359756

facts analysis examines, transforms, and models record sets to uncover valuable data and help make higher selections. Machine mastering techniques allow computer systems to automatically learn from records to make predictions primarily based on previously discovered patterns. By using these strategies, businesses can grow actual-time records analysis accuracy and see tendencies faster, allowing them to promptly respond to adjustments inside the surroundings. This paper will speak about using system studying techniques for actual-time statistics evaluation and the benefits of using such strategies. Mainly, techniques that include supervised studying, unsupervised learning, and deep studying will be discussed. Moreover, use instances in which system mastering may be applied, including economic fraud detection software programs, digital advertising and marketing, and predictive preservation, will be explored. Eventually, the outcomes of using system mastering strategies for data analysis, including statistics privateness and safety issues, can be mentioned. © 2023 IEEE.

关键词： Marketing

来源：评论

学校读者我要写书评

暂无评论

AutoSW: A new automated sliding window-based change point detection method for sensor data

AutoSW: A new automated sliding window-based change point de...

引用

2022 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2022

作者： Nejad, Ebrahim Behrouzian Silva, Carla Rodrigues, Arlete Jorge, Alipio Dutra, Ines University of Porto Faculty of Sciences Department of Computer Science Porto Portugal Bosch Security Systems - Sistemas de Segurança S.A FIR Software Engineering Ovar Ovar Portugal University of Porto Faculty of Sciences LIAAD and Department of Computer Science Porto Portugal University of Porto Faculty of Sciences Department of Computer Science and CINTESIS Porto Portugal

ISBN: (数字)9781665451260

ISBN: (纸本)9781665451260

Change point detection methods try to find any sudden changes in the patterns and features of a given time series. In this paper a new change point detection method is presented, where the window width is automatically calculated. The proposed algorithm, AutoSW, is based on a Sliding Window search method of the Python ruptures package and uses a subset of statistical concepts to compute a possibly optimal window width. The proposed algorithm is compared with some other popular methods such as PELT using different real-world and synthetic time series. Results show that AutoSW can perform better than PELT producing a better set of change points in the time series tested. © 2022 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Wind Driven Optimization-Based Medical Image Encryption for Blockchain-Enabled Internet of Things Environment

引用

computers, Materials & Continua 2022年第11期73卷 3219-3233页

作者： C.S.S.Anupama Raed Alsini N.Supriya E.Laxmi Lydia Seifedine Kadry Sang-Soo Yeo Yongsung Kim Department of Electronics&Instrumentation Engineering V.R.Siddhartha Engineering CollegeVijayawada520007India Department of Information Systems Faculty of Computing and Information TechnologyKing Abdulaziz UniversityJeddahSaudi Arabia Department of Computer Science and Engineering Malla Reddy Engineering College(A)Hyderabad500100India Department of Computer Science and Engineering Vignan’s Institute of Information TechnologyVisakhapatnam530049India Department of Applied Data Science Noroff University College4612KristiansandNorway Department of Computer Engineering Mokwon UniversityDaejeon35349Korea Department of Technology Education Chungnam National University 99Daehak-roYuseong-guDaejeon34134Korea

Internet of Things(IoT)and blockchain receive significant interest owing to their applicability in different application areas such as healthcare,finance,transportation,*** image security and privacy become a critical part of the healthcare sector where digital images and related patient details are communicated over the public *** paper presents a new wind driven optimization algorithm based medical image encryption(WDOA-MIE)technique for blockchain enabled IoT *** WDOA-MIE model involves three major processes namely data collection,image encryption,optimal key generation,and data ***,the medical images were captured from the patient using IoT ***,the captured images are encrypted using signcryption *** addition,for improving the performance of the signcryption technique,the optimal key generation procedure was applied by WDOA *** goal of the WDOA-MIE algorithm is to derive a fitness function dependent upon peak signal to noise ratio(PSNR).Upon successful encryption of images,the IoT devices transmit to the closest server for storing it in the blockchain *** performance of the presented method was analyzed utilizing the benchmark medical image *** security and the performance analysis determine that the presented technique offers better security with maximum PSNR of 60.7036 dB.

关键词： Internet of things image security medical images encryption optimal key generation blockchain

来源：评论

学校读者我要写书评

暂无评论

FAST CHARGING ELECTRIC BUSES IN MARIBOR: A COMPARATIVE ANALYSIS OF SIMULATION AND REAL-WORLD RESULTS 12th

FAST CHARGING ELECTRIC BUSES IN MARIBOR: A COMPARATIVE ANALY...

引用

12th International Conference on Environmental engineering, ICEE 2023

作者： Klemenčič, Mitja Španer, Marijan Lep, Marjan Moharić, Matej Department of Transportation Engineering Faculty of Civil Engineering Transportation Engineering and Architecture University of Maribor Smetanova 17 Maribor Slovenia Laboratory for Power Electronics Institute of Robotics Faculty of Electrical Engineering and Computer Science University of Maribor Koroška cesta 46 Maribor Slovenia

This paper presents a study of the energy consumption of a full electric bus charged at a fast-charging station with pantographs in the city of Maribor. The results of simulated and real tests on the PT line 6 are compared and analysed. The partly electrified PT line 6 provides insights on comparison between simulated energy consumption and re-al-life energy consumption, charging time and reliability of charging. The results reveal that the simulation considered several important factors like real life turnaround time, driving cycle, regenerative braking, charging time, peak hour charging time, heating/cooling systems, full passenger load as important and critical to identify the appropriate concept for charging. Technical and operational errors were identified in real world, due to connectivity failures or driver errors in positioning the vehicle, which were not considered in the simulation. The real-world data and procedure for selection of the most appropriate charging technology represent a good basis for full electrified Public Transport in Maribor. © 2023 The Author(s). Published by Vilnius Gediminas Technical University.

关键词： Electric buses

来源：评论

学校读者我要写书评

暂无评论

Text Mining for Intelligent Information Processing and Analysis 3

Text Mining for Intelligent Information Processing and Analy...

引用

3rd International Conference on Smart Generation Computing, Communication and Networking, SMART GENCON 2023

作者： Srivastav, Adarsh Indira, G. Tiwaskar, Shweta A. Garg, Avni Singh, Shweta Das, Sunanda Vivekananda Global University Department of Computer Science & Engineering Jaipur India Prince Shri Venkateshwara Padmavathy Engineering College Department of Electrical and Electronics Engineering Chennai127 India Vishwakarma Institute of Information Technology Department of Computer Engineering Pune India Centre of Interdisciplinary Research in Business and Technology Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India Faculty of Engineering and Technology Department of Computer Science and Engineering Karnataka Bangalore India

ISBN: (纸本)9798350319125

textual content Mining is the process of extracting meaningful information from large volumes of unstructured text. it's far a form of synthetic intelligence used to investigate and interpret files along with consumer feedback, emails, transcripts, net logs, surveys, and on line documents. Its intention is to discover hidden shape, patterns, institutions, and correlations in textual facts in order to draw inferences and broaden new insights. textual content Mining allows a deeper level of intelligence and selection-making by means of uncovering which means from unstructured textual facts. by means of utilising present day statistical strategies and herbal Language Processing (NLP), text Mining can produce automatic fashions that interpret and translate textual content into intelligent information which can be used to make correct predictions and provide actionable insights. textual content Mining is also being used to find out new trends, categorize client sentiment, locate correlations in purchaser comments, and section document collections. eventually, text Mining is being used to assist automate selection-making by automating workflows, processing massive volumes of text in a fragment of time, and providing deep insights into patron conversations. © 2023 IEEE.

关键词： Automating Categorize Collections Conversations Insights

来源：评论

学校读者我要写书评

暂无评论

Energy efficient indoor localisation for narrowband internet of things

引用

CAAI Transactions on Intelligence Technology 2023年第4期8卷 1150-1163页

作者： Ismail Keshta Mukesh Soni Mohammed Wasim Bhatt Azeem Irshad Ali Rizwan Shakir Khan Renato RMaaliw III Arsalan Muhammad Soomar Mohammad Shabaz Computer Science and Information Systems Department College of Applied SciencesAlMaarefa UniversityRiyadhSaudi Arabia Department of CSE University Centre for Research and Development Chandigarh UniversityMohaliPunjabIndia Department of Computer Science and Engineering National Institute of TechnologySrinagarIndia Asghar Mall College Rawalpindi Higher Education Department(HED)Govt.of the PunjabRawalpindiPakistan Department of Industrial Engineering Faculty of EngineeringKing Abdulaziz UniversityJeddahSaudi Arabia College of Computer and Information Sciences Imam Mohammad Ibn Saud Islamic University(IMSIU)RiyadhSaudi Arabia College of Engineering Southern Luzon State UniversityLucbanQuezonPhilippines Department of Automation Electronics and Electrical EngineeringGdansk University of TechnologyGdańskPoland Arba Minch University Arba MinchEthiopia

There are an increasing number of Narrow Band IoT devices being manufactured as the technology behind them develops *** high co‐channel interference and signal attenuation seen in edge Narrow Band IoT devices make it challenging to guarantee the service quality of these *** maximise the data rate fairness of Narrow Band IoT devices,a multi‐dimensional indoor localisation model is devised,consisting of transmission power,data scheduling,and time slot scheduling,based on a network model that employs non‐orthogonal multiple access via a *** on this network model,the optimisation goal of Narrow Band IoT device data rate ratio fairness is first established by the authors,while taking into account the Narrow Band IoT network:The multidimensional indoor localisation optimisation model of equipment tends to minimize data rate,energy constraints and EH relay energy and data buffer constraints,data scheduling and time slot *** a result,each Narrow Band IoT device's data rate needs are met while the network's overall performance is *** investigate the model's potential for convex optimisation and offer an algorithm for optimising the distribution of multiple resources using the KKT *** current work primarily considers the NOMA Narrow Band IoT network under a single EH ***,the growth of Narrow Band IoT devices also leads to a rise in co‐channel interference,which impacts NOMA's performance *** simulation,the proposed approach is successfully *** improvements have boosted the network's energy efficiency by 44.1%,data rate proportional fairness by 11.9%,and spectrum efficiency by 55.4%.

关键词： artificial inteligence detection of moving objects internet of things

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis on Tweets with Deep Learning and Natural Language Processing

Sentiment Analysis on Tweets with Deep Learning and Natural ...

引用

2022 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing, COM-IT-CON 2022

作者： Kusum Panda, Supriya P. Manav Rachna International Institute of Research & Studies Faculty of Engineering and Technology Department of Computer Science & Engineering Haryana Faridabad India

ISBN: (数字)9781665496025

ISBN: (纸本)9781665496025

Today's social media users are very large, where everyone expresses opinions, comments, criticisms, and so on. The data provides valuable information to be able to help people or groups in opinion evaluation. The enormous quantity of information is impractical for humans to interpret and evaluate manually. The Sentiment Analysis (SA) is a task that has become very important for the scientific and business community. Thus, SA can be used to diagnose a population regarding specific issues. There are many machine learning techniques, Deep Learning (DL) is one of them that has a deeper architecture than other ML techniques in solving prediction and classification problems. Using DL Models and NLP, this work tries to categorise content on social media into positive, negative, and neutral groups as sentiments. While NLP is being used to extract features, which enhances accuracy and performance, other DL models for classification, such as CNN, RNN, and LSTM models, can synthesize or create acceptable sentiments with a voluminous data;however, they take longer to train. © 2022 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Motor Speed Control Based on Enhanced Indirect-Field-Oriented Control Using Convolutional Neural Network 49

Motor Speed Control Based on Enhanced Indirect-Field-Oriente...

引用

49th Annual Conference of the IEEE Industrial Electronics Society, IECON 2023

作者： Sepeeh, Muhamad Syazmie Zulkifli, Shamsul Aizam Yi, Sim Sy Chiu, Huang-Jen Universiti Tun Hussein Onn Malaysia Faculty of Electrical and Electronics Engineering Department of Electrical Power Engineering Johor Malaysia National Taiwan University of Science and Technology Department of Electronic and Computer Engineering Taipei Taiwan

ISBN: (纸本)9798350331820

In this study, an indirect-field-oriented control (IFOC) of an induction motor (IM) was simulated and its speed was estimated using the model reference adaptive system (MRAS) topology of the modified voltage-model-based rotor-flux-oriented current model. As this study dealt with the convolutional neural network (CNN) implemented as the speed controller for the outer-loop control, the inner-loop control used proportional-integral (PI) control to control the current that was fed to the space vector pulse-width modulation (SVPWM) algorithm. The CNN used two inputs based on speed error, and it was designed and trained offline by employing the backpropagation (BP) algorithm to minimize the error between the desired speed and the speed estimated by the estimator and to improve the model's speed-tracking capability when there was an unpredicted speed at the input. The study also used a variable speed trajectory for testing with various speed references to mimic road speed conditions. The results are in good agreement with those of similar studies, and a comparison between the performances of the PI controller and the proposed CNN controller indicated that the CNN controller yielded a superior performance on IM speed. © 2023 IEEE.

关键词： Induction motors

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：