检索结果-内蒙古大学图书馆

applied machine learning for Information Security

DIGITAL THREATS: RESEARCH AND PRACTICE 2024年第1期5卷 1-5页

作者： Samtani, Sagar Raff, Edward Anderson, Hyrum Indiana Univ 1275 E 10th St Bloomington IN 47405 USA Booz Allen Hamilton 8283 Greensboro Dr Mclean VA 22102 USA Robust Intelligence 555 19th St San Francisco CA 94107 USA

Information security has undoubtedly become a critical aspect of modern cybersecurity practices. Over the past half-decade, numerous academic and industry groups have sought to develop machine learning, deep learning, and other areas of artificial intelligence-enabled analytics into information security practices. The Conference on applied machine learning (CAMLIS) is an emerging venue that seeks to gather researchers and practitioners to discuss applied and fundamental research on machine learning for information security applications. In 2021, CAMLIS partnered with ACM Digital Threats: Research and Practice (DTRAP) to provide opportunities for authors of accepted CAMLIS papers to submit their research for consideration into ACM DTRAP via a Special Issue on applied machine learning for Information Security. This editorial summarizes the results of this Special Issue.

关键词： applied machine learning deep learning artificial intelligence information security cybersecurity

来源：评论

学校读者我要写书评

暂无评论

applied machine learning for stroke differentiation by electrical impedance tomography with realistic numerical models

引用

BIOMEDICAL PHYSICS & ENGINEERING EXPRESS 2024年第1期10卷 015012-015012页

作者： Culpepper, Jared Lee, Hannah Santorelli, Adam Porter, Emily Univ Texas Austin Austin TX 78712 USA

Electrical impedance tomography (EIT) may have potential to overcome existing limitations in stroke differentiation, enabling low-cost, rapid, and mobile data collection. Combining bioimpedance measurement technologies such as EIT with machine learning classifiers to support decision-making can avoid commonly faced reconstruction challenges due to the nonlinear and ill-posed nature of EIT imaging. Therefore, in this work, we advance this field through a study integrating realistic head models with clinically relevant test scenarios, and a robust architecture consisting of nested cross-validation and principal component analysis. Specifically, realistic head models are designed which incorporate the highly conductive layers of cerebrospinal fluid in the subarachnoid space and ventricles. In total, 135 unique models are created to represent a large patient population, with normal, haemorrhagic, and ischemic brains. Simulated EIT voltage data generated from these models are used to assess the classification performance of support vector machines. Parameters explored include driving frequency, signal-to-noise ratio, kernel function, and composition of binary classes. Classifier accuracy at 60 dB signal-to-noise ratio, reported as mean and standard deviation, are (79.92% +/- 10.82%) for lesion differentiation, (74.78% +/- 3.79%) for lesion detection, (77.49% +/- 15.90%) for bleed detection, and (60.31% +/- 3.98%) for ischemia detection (after ruling out bleed). The results for each method were obtained with statistics from 3 independent runs with 17,280 observations, polynomial kernel functions, and feature reduction of 76% by PCA (from 208 to 50 features). While results of this study show promise for stroke differentiation using EIT data, our findings indicate that the achievable accuracy is highly dependent on the classification scenario and application-specific classifiers may be necessary to achieve acceptable accuracy.

关键词： anatomical modeling applied machine learning EIT stroke classification stroke detection bioimpedance measurement

来源：评论

学校读者我要写书评

暂无评论

Classifiable Limiting Mass Change Detection in a Graphene Resonator Using applied machine learning

引用

ACS applied ELECTRONIC MATERIALS 2022年第11期4卷 5184-5190页

作者： Seo, Miri Yang, Eunseo Shin, Dong Hoon Je, Yugyeong Joo, Chirlmin Lee, Kookjin Lee, Sang Wook Ewha Womans Univ Dept Phys Seoul 03760 South Korea Ewha Womans Univ Dept Sci & Engn Seoul 03760 South Korea Delft Univ Technol Kavli Inst Nanosci NL-2628 CJ Delft Netherlands Intel Corp Logic Technol Dev Qual & Reliabil Hillsboro OR 97124 USA

Nanomechanical resonator devices are widely used as ultrasensitive mass detectors for fundamental studies and practical applications. The resonance frequency of the resonators shifts when a mass is loaded, which is used to estimate the mass. However, the shift signal is often blurred by the thermal noise, which interferes with accurate mass detection. Here, we demonstrate the reduction of the noise interference in mass detection in suspended graphene-based nanomechanical resonators, by using applied machine learning. Featurization is divided into image and sequential datasets, and those datasets are trained and classified using 2D and 1D convolutional neural networks (CNNs). The 2D CNN learning-based classification shows a performance with f1-score over 99% when the resonance frequency shift is more than 2.5% of the amplitude of the thermal noise range.

关键词： applied machine learning deep learning graphene mass detection resonator

来源：评论

学校读者我要写书评

暂无评论

Searching String in Big-Data: A Better Approach by applied machine learning

引用

SN Computer Science 2021年第3期2卷 1-7页

作者： Singh, Paras Nath Gowdar, Tara P. Department of CSE CMR Institute of Technology AECS Layout Kundalahalli Bangalore 560037 India VTU Research Centre CMR Institute of Technology Bangalore 560037 India

There are hot patterns of huge data sets in numerous regions in the course of the most recent 5 years. Looking through string or pattern from a gigantic record is intense generally in the event of randomized situation. These require the advancement of new algorithmic strategies to investigate such enormous data sets and solve optimization errands utilizing sorting on indexing levels and applied machine learning Models. One of the essential strides in text processing is string searching and pattern matching. A word search algorithm works by finding the first or all the occurrences of a word in a textual data or ASCII files. The pre-processing phase is used to determine the formula for number of positions by which the pattern needs to be shifted in case of a mismatch in the matching phase. The fundamental objective of string search or pattern matching algorithms is to increase efficiency by reducing the number of comparisons and increase the length of shifts in event of a mismatch. The issue of efficiency of string search algorithms has probably never been considered so seriously and genuinely until the virtual content explosion caused by the web and the task of mining valuable data and information from it. In this paper, a better search algorithm “Tara–Paras String Search” is introduced that is faster than conventional Binary Search and Interpolation Search. Indexing levels are introduced by length of the word, sequence total of alphabets and starting letter of the word to reduce the size of input. For analysis, 2 data sets are considered. The dictionary of English words having more than 109,000 words and a list of more than 2.5 Lac sorted numbers and uniformly distributed (multiples of 7) are taken for data sets. Analysis and Implementations Models have been implemented, compared and executed in Python with time complexity and obviously applied machine learning will select the faster one. © 2021, The Author(s), under exclusive licence to Springer Nature Singapore Pt

关键词： applied machine learning Big-data Binary search Dictionary data structure Indexing Tara–Paras search

来源：评论

学校读者我要写书评

暂无评论

Guidelines for applied machine learning in construction industry-A case of profit margins estimation

引用

ADVANCED ENGINEERING INFORMATICS 2020年 43卷 101013-101013页

作者： Bilal, Muhammad Oyedele, Lukumon O. Univ West England Bristol Business Sch Big Data Analyt & Artificial Intelligence Lab BDA Bristol Avon England

The progress in the field of machine learning (ML) has enabled the automation of tasks that were considered impossible to program until recently. These advancements today have incited firms to seek intelligent solutions as part of their enterprise software stack. Even governments across the globe are motivating firms through policies to tape into ML arena as it promises opportunities for growth, productivity and efficiency. In reflex, many firms embark on ML without knowing what it entails. The outcomes so far are not as expected because the ML, as hyped by tech firms, is not the silver bullet. However, whatever ML offers, firms urge to capitalise it for their competitive advantage. Applying ML to real-life construction industry problems goes beyond just prototyping predictive models. It entails intensive activities which, in addition to training robust ML models, provides a comprehensive framework for answering questions asked by construction folks when intelligent solutions are getting deployed at their premises to substitute or facilitate their decision-making tasks. Existing ML guidelines used in the IT industry are vastly restricted to training ML models. This paper presents guidelines for applied machine learning (AML) in the construction industry from training to operationalising models, which are drawn from our experience of working with construction folks to deliver Construction Simulation Tool (CST). The unique aspect of these guidelines lies not only in providing a novel framework for training models but also answering critical questions related to model confidence, trust, interpretability, bias, feature importance and model extrapolation capabilities. Generally, ML models are presumed black boxes;hence argued that nobody knows what a model learns and how it generates predictions. Even very few ML folks barely know approaches to answer questions asked by the end users. Without explaining the competence of ML, the broader adoption of intelligent solutions

关键词： applied machine learning Profit margin forecasting Construction simulation tool Interpretable machine learning Predictive modelling

来源：评论

学校读者我要写书评

暂无评论

Advanced machine learning for missing petrophysical property imputation applied to improve the characterization of carbonate reservoirs

GEOENERGY SCIENCE AND ENGINEERING

引用

GEOENERGY SCIENCE AND ENGINEERING 2024年 238卷

作者： Abdulkhaleq, Hussein B. Khalil, Khalil A. Al-Mudhafar, Watheq J. Wood, David A. Basra Univ Oil & Gas Basrah Iraq Basrah Oil Co Basrah Iraq DWA Energy Ltd Lincoln England

Missing data is a common cause of uncertainty in reservoir characterization, especially in core analysis of porosity and permeability. This is due to the high cost and time required to gather core samples while drilling through the entire reservoir thickness. Therefore, the ability to collect core samples and directly assess petrophysical parameters is restricted by the small proportion of the reservoir thickness that can be measured by the retrieved cores. As a result, the absence of data has the potential to lead to inconsistent and imprecise classification of reservoirs and geomodeling, thereby increasing reservoir uncertainty. Therefore, imputation algorithms are needed to estimate petrophysical properties for incomplete intervals. This study comprehensively compares seven imputation techniques for predicting missing horizontal and vertical core permeability and core porosity data in a well drilled through a carbonate reservoir in a southern Iraqi gas field. The performance of each method was assessed using relative bias (RB) and reasonable descriptive statistical distributions (histogram, log view, and box plots of data ratios to compare the imputed and original data), and robustness to outliers (RtO). The results revealed that the random imputation of missing data (RIMD) technique is the most effective algorithm in addressing the 30.5% missing data (vertical permeability), resulting in the attainment of the RB value closest to zero (0.04076191) compared to other algorithms. Principal component analysis (PCA) and random forest (RF) yielded the most favorable results in terms of handling missing values and outliers for the 15% missing ratio dataset (core porosity and horizontal permeability). The advantages of these approaches are related to their capacity to independently and accurately impute the missing data without requiring extra information, such as well logging records. Additionally, the multiple imputation and machine learning algorithms fill missing val

关键词： Missing data imputation Core porosity and permeability Multiple imputation algorithms applied machine learning Descriptive statistics Carbonate reservoirs

来源：评论

学校读者我要写书评

暂无评论

Influence of Optimization Algorithms and Computational Complexity on Concrete Compressive Strength Prediction machine learning Models for Concrete Mix Design

引用

MATERIALS 2025年第6期18卷 1386-1386页

作者： Ziolkowski, Patryk Gdansk Univ Technol Fac Civil & Environm Engn Gabriela Narutowicza 11-12 PL-80233 Gdansk Poland

The proper design of concrete mixtures is a critical task in concrete technology, where optimal strength, eco-friendliness, and production efficiency are increasingly demanded. While traditional analytical methods, such as the Three Equations Method, offer foundational approaches to mix design, they often fall short in handling the complexity of modern concrete technology. machine learning-based models have demonstrated notable efficacy in predicting concrete compressive strength, addressing the limitations of conventional methods. This study builds on previous research by investigating not only the impact of computational complexity on the predictive performance of machine learning models but also the influence of different optimization algorithms. The study evaluates the effectiveness of three optimization techniques: the Quasi-Newton Method (QNM), the Adaptive Moment Estimation (ADAM) algorithm, and Stochastic Gradient Descent (SGD). A total of forty-five deep neural network models of varying computational complexity were trained and tested using a comprehensive database of concrete mix designs and their corresponding compressive strength test results. The findings reveal a significant interaction between optimization algorithms and model complexity in enhancing prediction accuracy. Models utilizing the QNM algorithm outperformed those using the ADAM and SGD in terms of error reduction (SSE, MSE, RMSE, NSE, and ME) and increased coefficient of determination (R2). These insights contribute to the development of more accurate and efficient AI-driven methods in concrete mix design, promoting the advancement of concrete technology and the potential for future research in this domain.

关键词： applied machine learning concrete concrete mix design concrete strength prediction data mining sustainable buildings sustainability construction industry innovation cement green building sustainable development

来源：评论

学校读者我要写书评

暂无评论

learning Location From Shared Elevation Profiles in Fitness Apps: A Privacy Perspective

引用

IEEE TRANSACTIONS ON MOBILE COMPUTING 2024年第1期23卷 581-596页

作者： Meteriz-Yildiran, Ulku Yildiran, Necip Fazil Kim, Joongheon Mohaisen, David Meta Menlo Pk CA 94025 USA Google Mountain View CA 94043 USA Korea Univ Dept Elect Engn Seoul 02841 South Korea Univ Cent Florida Dept Comp Sci Orlando FL 32816 USA

The extensive use of smartphones and wearable devices has facilitated many useful applications. For example, with Global Positioning System (GPS)-equipped smart and wearable devices, many applications can gather, process, and share rich metadata, such as geolocation, trajectories, elevation, and time. For example, fitness applications, such as Runkeeper and Strava, utilize the information for activity tracking and have recently witnessed a boom in popularity. Those fitness tracker applications have their own web platforms and allow users to share activities on such platforms or even with other social network platforms. To preserve the privacy of users while allowing sharing, several of those platforms may allow users to disclose partial information, such as the elevation profile for an activity, which supposedly would not leak the location of the users. In this work, and as a cautionary tale, we create a proof of concept where we examine the extent to which elevation profiles can be used to predict the location of users. To tackle this problem, we devise three plausible threat settings under which the city or borough of the targets can be predicted. Those threat settings define the amount of information available to the adversary to launch the prediction attacks. Establishing that simple features of elevation profiles, e.g., spectral features, are insufficient, we devise both natural language processing (NLP)-inspired text-like representation and computer vision-inspired image-like representation of elevation profiles, and we convert the problem at hand into text and image classification problem. We use both traditional machine learning- and deep learning-based techniques and achieve a prediction success rate ranging from 59.59% to 99.80%. The findings are alarming, highlighting that sharing elevation information may have significant location privacy risks.

关键词： Location privacy privacy breach privacy in social media fitness applications natural language processing applied machine learning

来源：评论

学校读者我要写书评

暂无评论

Approaches and Opportunities of Using machine learning Methods in Telecommunications and Industry 4.0

引用

MOBILE NETWORKS & APPLICATIONS 2024年第2期29卷 533-544页

作者： Cvitic, Ivan Jevremovic, Aleksandar Lameski, Petre Univ Zagreb Fac Transport & Traff Sci Vukeliceva 4 Zagreb 10000 Croatia Singidunum Univ Belgrade Serbia Univ Sts Cyril & Methodius Skopje Fac Comp Sci & Engn Skopje North Macedonia

Artificial intelligence can be considered a leading component of industrial transformation, but its application is also present in other areas, such as the telecommunications sector. Methodologies based on artificial intelligence, the most significant of which is machine learning, support these areas when predicting maintenance needs and reducing downtime. machine learning has many algorithms and methods of its own. This research presents machine learning methods and the possibilities of their use in the areas of telecommunications and Industry 4.0. With the improvement of new technologies such as higher internet speeds and 5G mobile networks in the field of telecommunications or Industry 4.0, comes the need for new and improved management and support for systems that use these new technologies. The possibilities of current Artificial Intelligence of Things devices and distributed execution of ML algorithms using Edge/Fog/Cloud computing environments are being considered. Some types of machine learning can be used for collecting data to improve user's quality of service. Also, some types of them can be used to collect data on network traffic or, in general, for any system that needs to collect data, cluster data points, and analyze data.

关键词： Artificial Intelligence Supervised machine learning applied machine learning Internet of Things (IoT) Artificial Intelligence of Things (AIoT) Edge computing Fog computing Cloud Distributed Denial of Service (DDoS)

来源：评论

学校读者我要写书评

暂无评论

machine-learning-Based Analysis of the Impact of Cylinder-Liner Rotation on Wear Rate in Internal Combustion Engines

引用

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING 2025年 1-15页

作者： Elrushaidat, Dana Farrukh, Habiba Alrousan, Ammar Jordan Univ Sci & Technol Dept Comp Sci Irbid Jordan Univ Calif Irvine Dept Comp Sci Irvine CA USA Yarmouk Univ Dept Ind Engn Irbid Jordan

The cylinder is the most expensive part of the Internal Combustion Engine in a cylinder-piston system. Friction along the cylinder liner results from the moving parts when the compression and intake strokes occur. Minimizing the abrasion on the cylinder wall is an ongoing research. The reduction of abrasion maintains the cylinder liner and expands its life span, which maintains the engine's thermal efficiency and reduces emissions resulting from the abrasion. This research investigates how cylinder rotation affects cylinder-liner friction through regression analysis. Our goal is to evaluate the effectiveness of cylinder rotation in minimizing abrasion on the cylinder liner by predicting and comparing friction levels between conventional fixed cylinders and rotating cylinders. We employ machine learning tools to explore the optimal rotation angle for the rotating cylinder. Prediction of wear rate aids in scheduling preventive maintenance for rotating and fixed cylinders to maintain engine efficiency, vehicle fuel economy, and emission standards. Due to the experimental dataset's limited size and data collection difficulties, we use augmentation and feature scaling techniques to address these challenges. The decision-tree regression model yields a predictive accuracy of R2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R<^>2$$\end{document} = 0.995 and a normalized root mean square error of 0.006, indicating high precision in its predictions. Our prediction model accurately determines the optimal rotation angle based on engine speed. Additionally, it assists in predicting maintenance needs for both fixed and rotating cylinder systems. The decision-tree prediction model offers compelling evidence that implementing a rotating cylinder can extend the cylinder's lifespan four times compared to

关键词： Cylinder wear Rotational cylinder Internal combustion engine Artificial intelligence applied machine learning Linear regression Polynomial regression Support vector machine (SVM) Radial basis function (RBF) Multilayer perceptron (MLP) regression Decision tree (DT) regression

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：