检索结果-内蒙古大学图书馆

supervised and unsupervised learning using a fully-plastic all-optical unit of artificial intelligence based on solitonic waveguides

引用

NEURAL COMPUTING & APPLICATIONS 2021年第24期33卷 17071-17079页

作者： Bile, Alessandro Moratti, Francesca Tari, Hamed Fazio, Eugenio Sapienza Univ Roma Dept Fundamental & Appl Sci Engn Via A Scarpa 16 Rome Italy

The software implementations of neuronal systems have shown great effectiveness, even if the natural hardware separation between the processing and memory areas in computers slows down the analysis capacity. To overcome these limitations, new hardware configurations are moving towards neuromorphic models, capable of unifying the processing/memory dichotomy. Recently, integrated photonic X-junctions formed by waveguides written by spatial solitons have shown the ability to perform supervised learning. The solitonic technology, compared to the traditional one, offers the advantage of realizing plastic circuitry, a typical characteristic of biological neural networks. This work extensively studies both supervised and unsupervised learning of photonic soliton X-junctions. By exploiting the plasticity of the nonlinear refractive index at the base of the soliton formation, X-junctions can readdress their behaviours forwarding data to different outputs. In this article, we will extend the state-of-the-art: starting from supervised learning, for which all possible cases are now investigated, a material sensitive to the transported signals will be introduced to allow the junction to carry out unsupervised learning. In this way, the junction autonomously recognises the transported signals without the external intervention of the operator. learning and memory now physically coincide in fact, learning means that the junction slowly switches based on the information sent;any further unknown information sent will find the junction in the modified state which corresponds to the learned information and will be recognised as well (reasoning based on comparison with stored information).

关键词： Neuromorphic supervised and unsupervised learning Solitons All-optical intelligent system

来源：评论

学校读者我要写书评

暂无评论

A wavelet-based neural network scheme for supervised and unsupervised learning

引用

NEURAL COMPUTING & APPLICATIONS 2021年第20期33卷 13433-13448页

作者： Esquivel, Manuel L. Krasii, Nadezhda P. Univ Nova Lisboa Dept Math FCT NOVA Campus Caparica P-2829516 Monte De Caparica Portugal Don State Tech Univ Gagarin Sq 1 Rostov Na Donu 344000 Russia

We introduce a scheme for supervised and unsupervised learning based on successive decomposition of random inputs by means of wavelet basis. To each successive layer corresponds a different wavelet basis with more null moments than its predecessor. We consider two types of operations in the scheme: firstly, a input signal treatment phase-the awake phase-and, secondly, a reorganizing phase of the random wavelet coefficients obtained in the previous awake phase-the asleep phase. The next awake phase input treatment will include a feedback derived from the previous asleep phase. The set of random wavelet coefficients of the deepest layer-at each stage of the learning process-is supposed to be Gaussian distributed, and the corresponding sequence of Gaussian distributions constitutes the inner representation of the world in the scheme. We show that in the case of a constant average value of the inputs in each successive awake phases, the mean value of the feedback converges to a multiple of the constant expected value of the inputs. We show a general result on the stabilization of the Gaussian distribution corresponding to the deepest layer, and we show that when the estimated means and covariances converge, then the sequence of Gaussian distributions of the inner representation of the world in the scheme also converges. We present an example of a neural network scheme for supervised learning corresponding to extracting a signal from data corrupted with Gaussian noise.

关键词： Neural networks supervised and unsupervised learning Wavelet theory

来源：评论

学校读者我要写书评

暂无评论

Joint supervised and unsupervised Machine learning for Spaceborne Spectrum Sensing

Joint Supervised and Unsupervised Machine Learning for Space...

引用

2024 International Conference on Wireless for Space and Extreme Environments

作者： Manzoor, Bisma Al-Hourani, Akram RMIT Univ Sch Engn Melbourne Vic Australia

ISBN: (纸本)9798350351118;9798350351125

As Internet of Things (IoT) applications continue to expand, the increasing traffic in frequency spectrum highlights critical challenges related to spectrum utilization, IoT device management, and potential security risk from undetected devices. As such, a comprehensive visibility of IoT devices across the frequency spectrum is essential, especially given the ubiquity of these devices. While terrestrial networks provide local visibility, broader coverage can be achieved through spectrum sensing via non-terrestrial platforms, such as satellites. However, the complex nature of the satellite radio channel introduces additional signal losses, making traditional spectrum sensing techniques inadequate. As such, this work presents a novel methodology for detecting low power radio frames transmitted over a complex satellite channel. We implement joint machine learning comprising a supervised Semantic Segmentation neural network and unsupervised Density-Based Spatial Clustering of Applications with Noise (DBSCAN) to detect IoT packets from the spectrogram images. The proposed method achieves 92% mean accuracy and 93% mean Intersection-over-Union (IoU) at low Signal-to-Noise Ratio (SNR) level. Moreover, the proposed method outperforms the traditional threshold detection methodologies as demonstrated by the Receiver Operating Characteristics (ROC).

关键词： Space Spectrum Sensing Satellites Machine learning IoT Semantic Segmentation Deep learning SNR supervised and unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing Business with AI: Unlocking Customer Insights Through unsupervised and supervised learning for Behavior Prediction 9th

Revolutionizing Business with AI: Unlocking Customer Insight...

引用

9th International Congress on Information and Communication Technology (ICICT)

作者： Rachini, Ali Fares, Charbel Assaf, Maroun Abi Jaber, Mustafa Musa Holy Spirit Univ Kaslik 446 Jounieh Lebanon NH USA Informat Inst Postgrad Studies Iraqi Commiss Comp & Informat Baghdad Iraq

ISBN: (纸本)9789819732982;9789819732999

This study navigates the intricate landscape of understanding and classifying customer behavior amid dynamic consumer preferences and vast data. Conventional segmentation methods often fall short, resulting in suboptimal marketing strategies and reduced satisfaction. Our research advocates for an advanced approach integrating unsupervised and supervised learning techniques to enhance accuracy in customer behavior classification. By synergizing these methodologies, businesses can precisely tailor products and services, fostering engagement and success in the marketplace. The study emphasizes the critical need for a data-driven, machine learning-based solution to unlock the full potential of customer behavior analysis. We present a framework advancing customer behavior classification through unsupervised learning, encompassing data exploration, clustering, and dimensionality reduction. The research extends its impact by enhancing application security for sensitive data. Evaluations demonstrate outstanding accuracy levels for each algorithm: Logistic Regression (95%), Random Forest (99%), SVM (99%), CNN (99.3%), and ANN (99.4%). These results underscore the efficacy of our proposed methodology, representing a significant advancement in precision and depth of customer behavior classification.

关键词： E-commerce Customer behavior Predictive modeling supervised and unsupervised learning AI

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based supervised and unsupervised neural networks for analysing the characteristics of powder composite preforms

引用

INTERNATIONAL JOURNAL OF MODELLING AND SIMULATION 2021年第6期41卷 451-462页

作者： Pavanasam, Radha Chandrasekaran, G. Selvakumar, N. Mepco Schlenk Engn Coll Dept Comp Applicat Sivakasi 626005 India Mepco Schlenk Engn Coll Dept Mech Engn Sivakasi India

This highlights the importance of deep learning in analysing the characteristics of composite preforms in the manufacturing of mechanical components in the industries. The raw data of Powder Metallurgy Lab are having highly non-linear, noisy and interrelated data. To process this kind of data, both shallow and deep neural network models of supervised learning have been considered for predicting the properties of composites. This work proves that the deep forward neural networks have good generality in recognizing even 100% of independent unseen data. The composites under the manufacturing process may have pores, which will reduce the strength and life time of the materials. Hence the sintering and extrusion of material processes are being followed to reduce the strength of pores. The presence of pores are examined by generating Scanning Electron Microscope (SEM) images for Cu-(5-20%)W composite preforms with a density of 94% without destroying the materials before and after the sintering and extrusion process. The pores are analyzed using unsupervised neural network model with Deep learning paradigm. These deep-learning based supervised and unsupervised models will guide the Lab Engineers to avoid the expensive experimentation and risky environment while preparing sintered composite preforms.

关键词： Deep neural network computational intelligence powder metallurgy composite preforms pore analysis sem images supervised and unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Predicting Underwater Noise Spectra Dominated by Wind Turbine Contributions

引用

IEEE JOURNAL OF OCEANIC ENGINEERING 2024年第4期49卷 1675-1694页

作者： Trucco, Andrea Univ Genoa Dept Elect Elect Telecommun Engn & Naval Architect Genoa I-16145 Italy

The study of the impact on the marine ecosystem of an offshore wind farm benefits from the knowledge of the underwater noise observed at a single turbine, as the wind speed varies. The calculation of the noise spectral average at a given wind speed requires many recordings, each acquired in a limited time interval: an extremely time-consuming process. This study investigated how to approach the spectral average using only very few noise recordings for each wind speed, leveraging supervised and unsupervised machine learning techniques. Three different prediction methods, based on mean and interpolation, principal component analysis (PCA), and nonnegative matrix factorization, in combination with four techniques for coefficient estimation as the wind varies, are tested. Prediction based on principal component analysis, combined with Gaussian process regression, outperforms other methods in all three case studies considered. The latter, in addition to the problem described above, include the prediction of the noise spectrum: at wind speeds where no noise recordings are available, and using a few recordings acquired at another (nominally identical) wind turbine.

关键词： Noise Wind speed Principal component analysis Wind turbines Vectors Recording supervised learning Offshore wind turbine principal component analysis (PCA) spectral prediction supervised and unsupervised learning underwater noise

来源：评论

学校读者我要写书评

暂无评论

Most influential feature form for supervised learning in voltage sag source localization

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年 133卷

作者： Mohammadi, Younes Polajzer, Bostjan Leborgne, Roberto Chouhy Khodadad, Davood Umea Univ Dept Appl Phys & Elect S-90187 Umea Sweden Univ Maribor Fac Elect Engn & Comp Sci Maribor 2000 Slovenia Univ Fed Rio Grande do Sul Ave Osvaldo Aranha 99 BR-90035190 Porto Alegre RS Brazil Haken Gullesons Vag 20 S-90187 Umea Sweden

The paper investigates the application of machine learning (ML) for voltage sag source localization (VSSL) in electrical power systems. To overcome feature -selection challenges for traditional ML methods and provide more meaningful sequential features for deep learning methods, the paper proposes three time -sample -based feature forms, and evaluates an existing feature form. The effectiveness of these feature forms is assessed using k -means clustering with k = 2 referred to as downstream and upstream classes, according to the direction of voltage sag origins. Through extensive voltage sag simulations, including noises in a regional electrical power network, kmeans identifies a sequence involving the multiplication of positive -sequence current magnitude with the sine of its angle as the most prominent feature form. The study develops further traditional ML methods such as decision trees (DT), support vector machine (SVM), random forest (RF), k -nearest neighbor (KNN), an ensemble learning (EL), and a designed one-dimensional convolutional neural network (1D -CNN). The results found that the combination of 1D -CNN or SVM with the most prominent feature achieved the highest accuracies of 99.37% and 99.13%, respectively, with acceptable/fast prediction times, enhancing VSSL. The exceptional performance of the CNN was also approved by field measurements in a real power network. However, selecting the best ML methods for deployment requires a trade-off between accuracy and real-time implementation requirements. The research findings benefit network operators, large factory owners, and renewable energy park producers. They enable preventive maintenance, reduce equipment downtime/damage in industry and electrical power systems, mitigate financial losses, and facilitate the assignment of power -quality penalties to responsible parties.

关键词： Voltage sag (dip) Source localization supervised and unsupervised learning Convolutional neural network Time -sample -based features

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Intrusion Detection Systems and Machine learning-Based Model Analysis Through Decision Tree

引用

IEEE ACCESS 2023年 11卷 80348-80391页

作者： Azam, Zahedi Islam, Md. Motaharul Huda, Mohammad Nurul United Int Univ Dept Comp Sci & Engn Dhaka 1212 Bangladesh

Cyber-attacks pose increasing challenges in precisely detecting intrusions, risking data confidentiality, integrity, and availability. This review paper presents recent IDS taxonomy, a comprehensive review of intrusion detection techniques, and commonly used datasets for evaluation. It discusses evasion techniques employed by attackers and the challenges in combating them to enhance network security. Researchers strive to improve IDS by accurately detecting intruders, reducing false positives, and identifying new threats. Machine learning (ML) and deep learning (DL) techniques are adopted in IDS systems, showing potential in efficiently detecting intruders across networks. The paper explores the latest trends and advancements in ML and DL-based network intrusion detection systems (NIDS), including methodology, evaluation metrics, and dataset selection. It emphasizes research obstacles and proposes a future research model to address weaknesses in the methodologies. The decision tree, known for its speed and user-friendliness, is proposed as a model for detecting result anomalies, combining findings from a comparative survey. This research aims to provide insights into building an effective decision tree-based detection framework.

关键词： Intrusion detection system machine learning inductive learning DDoS attacks decision tree supervised and unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Faults and fractures detection using a combination of seismic attributes by the MLP and UVQ artificial neural networks in an Iranian oilfield

引用

PETROLEUM SCIENCE AND TECHNOLOGY 2023年第24期41卷 2299-2327页

作者： Mohammadi, Reza Bakhtiari, Mohammad Reza Amirkabir Univ Technol Dept Petr Engn 3550 Hafez AveValiasr Sq Tehran *** Iran Natl Iranian Drilling Co NIDC Dept Well Logging Ahvaz Iran Natl Iranian Oil Co Explorat Directorate NIOC EXP Dept Geophys Tehran Iran

Faults and fractures play a significant role in oilfield drilling operations, hydrocarbon trapping, and reservoir development. Exploring faults quickly and accurately can help reach the target more efficiently. In this study, applicable seismic attributes were combined using a multilayer perceptron and an unsupervised vector quantizer and applied to a 3D seismic cube to identify discontinuities. First, high-probability faulted areas were picked manually on a seismic section as an input pattern for the MLP. Then, particular seismic attributes (dip-steering, similarity, coherency, curvature, instantaneous) were applied to the data. Consequently, the MLP and UVQ were used to determine the most contributed attributes. Using the MLP and UVQ, the most relevant attributes were integrated to find faults and fractures in the 3D seismic volume. In contrast to some fault-identifying methods and prior studies, this study used not just steered attributes but also compared supervised and unsupervised neural networks. Eventually, comparing the outputs of the MLP, faulted and non-faulted cubes, with the initial seismic section and the UVQ's output revealed discrepancies. For a specific set of attributes, the MLP was obviously superior to the UVQ in terms of creating detailed outputs, analyzing time, and rendering more precise and trustworthy results.

关键词： artificial neural network (ANN) coherency curvature dip-steering energy multilayer perceptron (MLP) seismic attributes similarity supervised and unsupervised learning unsupervised vector quantizer (UVQ)

来源：评论

学校读者我要写书评

暂无评论

Clustering of precipitation features by a new validity index over Iran

引用

JOURNAL OF HYDROLOGY-REGIONAL STUDIES 2025年 59卷

作者： Moghim, Sanaz Rajabi, Reza Sharif Univ Technol Dept Civil Engn Azadi Ave Tehran Iran

Study region: The study domain is Iran, a climatically diverse country. Study focus: Clustering is an effective approach for climate zoning that identifies areas with similar climatic patterns. This study clusters precipitation data over Iran to identify meaningful regional patterns based on six features, including extreme indices (Rx1day, SDII, R95p, R99p, R10) and monthly average precipitation (Pmon) during 2001-2020. To determine an optimal number of clusters, this study defines a modified validity index, SVM, by refining the separation measure used in the traditional SV index. Four traditional validity indices, including DB, Gap Statistic, Silhouette, and CH, are also used to evaluate the performance of the new index. New hydrological insights for the region: Results indicate that traditional clustering indices tend to either overestimate or underestimate the optimal number of clusters, which can be due to the complex pattern of precipitation clusters with no clearly defined boundaries. However, the developed index outperforms traditional indices on clustering synthetic datasets with varying sizes and shapes, and also on the real precipitation data. Results show that seven, five, seven, four, three, and six clusters of the desired features cover precipitation zones over Iran, which is consistent with the Koppen-Geiger Classification (KGC) and previous studies. In addition, results highlight the main role of the topography, like mountains, that affect air masses and circulation in precipitation zones clustering. The developed validity index can be efficiently used for clustering other hydrometeorological variables and climatic zones, which is valuable in water resources, atmosphere, and climate studies, and hazard assessment.

关键词： Clustering supervised and unsupervised learning Validity indices SV index Climate classification Precipitation

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：