检索结果-内蒙古大学图书馆

Benchmark of machine learning algorithms on capturing future distribution network anomalies

IET GENERATION TRANSMISSION & DISTRIBUTION 2019年第8期13卷 1441-1455页

作者： Mohammadpourfard, Mostafa Weng, Yang Tajdinian, Mohsen Sahand Univ Technol Dept Elect & Comp Engn Tabriz Iran Arizona State Univ Sch Elect Comp & Energy Engn Tempe AZ 85287 USA Shiraz Univ Sch Elect & Comp Engn Shiraz Iran

The conventional distribution network is undergoing structural changes and becoming an active grid due to the advent of smart grid technologies encompassing distributed energy resources (DERs), aggregated demand response and electric vehicles (EVs). This establishes a need for state estimation-based tools and real-time monitoring of the distribution grid to correctly apply active controls. Although such new tools may be vulnerable to cyber-attacks, cyber-security of distribution grid has not received enough attention. As smart distribution grid intensively relies on communication infrastructures, the authors assume in this study that an attacker can compromise the communication and successfully conduct attacks against crucial functions of the distribution management system, making the distribution system prone to instability boundaries for collapses. They formulate the attack detection problem in the distribution grid as a statistical learning problem and demonstrate a comprehensive benchmark of statistical learning methods on various IEEE distribution test systems. The proposed learning algorithms are tested using various attack scenarios which include distinct features of modern distribution grid such as integration of DERs and EVs. Furthermore, the interaction between transmission and distribution systems and its effect on the attack detection problem are investigated. Simulation results show attack detection is more challenging in the distribution grid.

关键词： active control Smart grids Network security machine learning algorithms BENCHMARKS Power grids dispensing system

来源：评论

学校读者我要写书评

暂无评论

Comprehensive and Empirical Evaluation of machine learning algorithms for Small Molecule LC Retention Time Prediction

引用

ANALYTICAL CHEMISTRY 2019年第5期91卷 3694-3703页

作者： Bouwmeester, Robbin Martens, Lennart Degroeve, Sven VIB UGent Ctr Med Biotechnol Ghent Belgium Univ Ghent Dept Biomol Med Ghent Belgium

Liquid chromatography is a core component of almost all mass spectrometric analyses of (bio)molecules. Because of the high-throughput nature of mass spectrometric analyses, the interpretation of these chromatographic data increasingly relies on informatics solutions that attempt to predict an analyte's retention time. The key components of such predictive algorithms are the features these are supplies with, and the actual machine learning algorithm used to fit the model parameters. Therefore, we have evaluated the performance of seven machine learning algorithms on 36 distinct metabolomics data sets, using two distinct feature sets. Interestingly, the results show that no single learning algorithm performs optimally for all data sets, with different types of algorithms achieving top performance for different types of analytes or different protocols. Our results thus show that an evaluation of machine learning algorithms for retention time prediction is needed to find a suitable algorithm for specific analytes or protocols. Importantly, however, our results also show that blending different types of models together decreases the error on outliers, indicating that the combination of several approaches holds substantial promise for the development of more generic, high-performing algorithms.

关键词： small molecule Core Assembly Retention Time machine learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Hydrogen production via biomass gasification, and modeling by supervised machine learning algorithms

引用

INTERNATIONAL JOURNAL OF HYDROGEN ENERGY 2019年第32期44卷 17260-17268页

作者： Ozbas, Emine Elmaslar Aksu, Dogukan Ongen, Atakan Aydin, Muhammed Ali Ozcan, H. Kurtulus Istanbul Univ Cerrahpasa Engn Fac Dept Environm Engn TR-34320 Istanbul Turkey Istanbul Univ Cerrahpasa Engn Fac Dept Comp Engn TR-34320 Istanbul Turkey

Prediction of clean hydrogen production via biomass gasification by supervised machine learning algorithms was studied. Lab-scale gasification studies were performed in a steel fixed bed updraft gasifier having a cyclone separator. Pure oxygen, and dried air with varying flow rates (0.05-0.3 L/min) were applied to produce syngas (H-2, CH4, CO). Gas compositions were monitored via on-line gas analyzer. Various regression models were created by using different machine learning (ML) algorithms which are Linear Regression (LR), K Nearest Neighbors (KNN) Regression, Support Vector machine Regression (SVMR) and Decision Tree Regression (DTR) algorithms to predict the value of H-2 concentration based on the other parameters that are time, temperature, CO, CO2, CH4, O-2 and heating value. The highest hydrogen value in syngas was found around 35% vol. after gasification experiments with higher heating value (HHV) of approximately 3400 kcal/m30.05 L/min and 0.015 L/min were the optimum flow rates for dried air and pure oxygen, respectively. In modeling section, it was observed that H2 concentrations were being reflected effectively by the concentrations estimated through the proposed model structures, and by having r(2) values of 0.99 which were ascertained between actual and model results. (C) 2019 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

关键词： machine learning algorithms Gasification Biomass Hydrogen

来源：评论

学校读者我要写书评

暂无评论

A comparison of machine learning algorithms and covariate balance measures for propensity score matching and weighting

引用

BIOMETRICAL JOURNAL 2019年第4期61卷 1049-1072页

作者： Cannas, Massimo Arpino, Bruno Univ Cagliari Dept Econ & Business Sci Viale St Ignazio 84 I-09123 Cagliari Italy Univ Firenze Dept Stat Comp Sci Applicat Florence Italy

Propensity score matching (PSM) and propensity score weighting (PSW) are popular tools to estimate causal effects in observational studies. We address two open issues: how to estimate propensity scores and assess covariate balance. Using simulations, we compare the performance of PSM and PSW based on logistic regression and machine learning algorithms (CART;Bagging;Boosting;Random Forest;Neural Networks;naive Bayes). Additionally, we consider several measures of covariate balance (Absolute Standardized Average Mean (ASAM) with and without interactions;measures based on the quantile-quantile plots;ratio between variances of propensity scores;area under the curve (AUC)) and assess their ability in predicting the bias of PSM and PSW estimators. We also investigate the importance of tuning of machine learning parameters in the context of propensity score methods. Two simulation designs are employed. In the first, the generating processes are inspired to birth register data used to assess the effect of labor induction on the occurrence of caesarean section. The second exploits more general generating mechanisms. Overall, among the different techniques, random forests performed the best, especially in PSW. Logistic regression and neural networks also showed an excellent performance similar to that of random forests. As for covariate balance, the simplest and commonly used metric, the ASAM, showed a strong correlation with the bias of causal effects estimators. Our findings suggest that researchers should aim at obtaining an ASAM lower than 10% for as many variables as possible. In the empirical study we found that labor induction had a small and not statistically significant impact on caesarean section.

关键词： caesarean section covariate balance labor induction machine learning algorithms propensity score methods

来源：评论

学校读者我要写书评

暂无评论

Efficient prediction of fog-related low-visibility events with machine learning and evolutionary algorithms

引用

ATMOSPHERIC RESEARCH 2023年 295卷

作者： Pelaez-Rodriguez, C. Perez-Aracil, J. Casanova-Mateo, C. Salcedo-Sanz, S. Univ Alcala Dept Signal Proc & Commun Alcala De Henares 28805 Spain Univ Politecn Madrid Dept Comp Syst Engn Madrid 28038 Spain

Low visibility events are a severe problem for road transport, causing accidents and major economic losses. Their accurate prediction may help prevent these problems. For that purpose, machine and deep learning techniques have been applied for fog prediction using in situ meteorological data and persistence variables as baseline predictors. These techniques have been evaluated for different prediction time-horizons: 1 h, 3 h and 6 h. The effect of including data extracted from ERA5 Reanalysis as predictive variables has been studied. A database, covering 23 months, has been used, which contains visibility and other meteorogical variables measured in Mondon similar to edo, Galicia, Spain. A 222000 km2 region around Mondon similar to edo has been delimited. Thus, a proposed iterative forward selection algorithm based on evolutionary algorithms has been applied to determine the optimal variables and nodes in the region for each regressor model. Both Differential Evolution and Particle Swarm Optimization have been used as optimization algorithms, and an improvement of up to 17.3% with respect to the baseline databases have been obtained. Finally, an analysis of the most frequently selected variables by the evolutionary algorithms has been conducted, leading us to conclude which variables and geographical nodes provide better information to the prediction models.

关键词： Low -visibility events Orographic fog machine learning algorithms Reanalysis data Evolutionary optimization algorithms Iterative forward selection

来源：评论

学校读者我要写书评

暂无评论

Big data and machine learning algorithms for health-care delivery (vol 20, pg e262, 2019)

引用

LANCET ONCOLOGY 2019年第6期20卷 E293-E293页

作者： Ngiam, K. Y. Khor, I. W. [a]Department of Surgery National University of Singapore Singapore [b]Department of Medicine Yong Loo Lin School of Medicine National University of Singapore Singapore [c]Division of General Surgery (Thyroid and Endocrine Surgery) University Surgical Cluster National University Hospital Singapore [d]National University Health System Corporate Office Singapore

Summary Analysis of big data by machine learning offers considerable advantages for assimilation and evaluation of large amounts of complex health-care data. However, to effectively use machine learning tools in health care, several limitations must be addressed and key issues considered, such as its clinical implementation and ethics in health-care delivery. Advantages of machine learning include flexibility and scalability compared with traditional biostatistical methods, which makes it deployable for many tasks, such as risk stratification, diagnosis and classification, and survival predictions. Another advantage of machine learning algorithms is the ability to analyse diverse data types (eg, demographic data, laboratory findings, imaging data, and doctors’ free-text notes) and incorporate them into predictions for disease risk, diagnosis, prognosis, and appropriate treatments. Despite these advantages, the application of machine learning in health-care delivery also presents unique challenges that require data pre-processing, model training, and refinement of the system with respect to the actual clinical problem. Also crucial are ethical considerations, which include medico-legal implications, doctors’ understanding of machine learning tools, and data privacy and security. In this Review, we discuss some of the benefits and challenges of big data and machine learning in health care.

关键词： machine learning algorithms big data

来源：评论

学校读者我要写书评

暂无评论

Investigation of machine learning-Based algorithms for Identification of Diabetes Disease

Investigation of Machine Learning-Based Algorithms for Ident...

引用

Human-Computer Interaction, Optimization and Robotic Applications (HORA), International Congress on

作者： Ali Shafique Arslan Saleem Abdul Sami Muhammad Hamza Ishaq Hazrat Bilal Yeliz Yesilada Computer Engineering Middle East Technical University Northern Cyprus Campus Kalkanli Turkiye Electrical and Electronics Engineering Middle East Technical University Northern Cyprus Campus Kalkanli Turkiye School of Automation Engineering University of Electronic Science and Technology of China Chengdu China School of Computer Science University of Galway Galway Ireland

ISBN: (数字)9798331510886

ISBN: (纸本)9798331510893

Diabetes, a condition characterized by disrupted insulin production or insulin resistance, results in elevated blood glucose levels. It can lead to severe issues such as heart disease and kidney failure. Early prediction of diabetes is crucial for mitigating risks and managing the severity of the condition. The paper presents effective algorithms of machine learning (ML) for diabetes prediction, including Decision Trees (DT), K-Nearest Neighbors (KNN), Light Gradient-Boosting machine (LGBM), Support Vector machine (SVM), and Random Forest (RF), all based on ML algorithms. Various ML models’ performance was assessed using the PID Indian Diabetes (PID) dataset, encompassing parameters like age, glucose levels, insulin levels, and skin thickness, among others. The examination unveiled that among all models, the LGBM algorithm showcased superior performance, attaining an impressive accuracy of 92%. Consequently, LGBM emerges as the most efficient algorithm for accurately distinguishing between diabetic and non-diabetic individuals.

关键词： Support vector machines Measurement machine learning algorithms Accuracy Nearest neighbor methods Prediction algorithms Diabetes Insulin Medical diagnostic imaging Diseases

来源：评论

学校读者我要写书评

暂无评论

Review on Traffic Prediction Framework Using machine and Deep learning algorithms

Review on Traffic Prediction Framework Using Machine and Dee...

引用

International Conference on ent Computing and Control Systems (ICICCS)

作者： R. Anto Pravin Nancy Noella R S Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Tamil Nadu India

ISBN: (数字)9798331512088

ISBN: (纸本)9798331512095

Management of road traffic and congestion is a critical feature of urban planning and infrastructure development. Traffic congestion has recently become a major problem on a global scale due to the rapid increase in industrialization and the number of vehicles on the road. Efficient traffic management not only enhances the flow of vehicles but also contributes to reducing fuel consumption, minimizing air pollution, and improving road safety. Traditional traffic management methods, which primarily rely on static models and historical data, have proven to be insufficient in handling the complexities of modern transportation systems. Therefore, there is a growing need for advanced techniques that can predict traffic patterns and optimize traffic flow in real time. Techniques of machine learning and deep learning have become increasingly effective in recent years for addressing the difficulties in controlling traffic and predictions. By influencing the vast amount of data generated from traffic sensors, GPS devices, and other sources, machine learning and deep learning algorithms can analyze, learn, and predict traffic conditions more accurately than traditional methods. These techniques are particularly effective in handling the dynamic and nonlinear nature of traffic flow, making them ideal for real-time applications in traffic control, congestion prediction, and incident management.

关键词： Deep learning machine learning algorithms Heuristic algorithms Transportation Traffic control Prediction algorithms Real-time systems Vehicle dynamics Traffic congestion Optimization

来源：评论

学校读者我要写书评

暂无评论

Early Detection of Chronic Diseases Using machine and Deep learning algorithms

Early Detection of Chronic Diseases Using Machine and Deep L...

引用

Sentiment Analysis and Deep learning (ICSADL), International Conference on

作者： Bhushan Fulkar Trishul Dhale Utkarsha Pacharaney Shruti Deshmukh Department of Artificial intelligence And Data science Datta Meghe institute of Higher Education and Research Wardha India

ISBN: (数字)9798331523923

ISBN: (纸本)9798331523930

Today's persons suffer from a wide variety of diseases due to various influences and choices made at the community level. Thus, to prevent the occurrence of such illnesses persistent identification and prediction are paramount. Manually determining the disorders is generally challenging for the doctors to be accurate with the exact numbers. The objective of this study is thus to use these chronic conditions and an accompanying test to select people who will be predominantly affected by these diseases. This might be done by making certain that this categorization correctly singles out people with chronic diseases employing the most advanced data learning technique. Another tough issue is a prediction concerning health conditions. It can therefore be concluded that data mining is an important component of sickness prediction. Employing automatic feature extraction and disease prediction by CNN, distance computation employing KNN for identification of the closest match or data within the data set, and the last disease prognosis of a broad disease, the proposed system enables the determination of patient's symptoms. The patient's life history, medical history of consultations, and illness symptoms were used to construct the data client set. It is on this comprehensive sickness prediction that all of these details were considered. The above said method is compared with logistic regression, Decision tree and Naive Bayes in the conclusion part of the paper.

关键词： Deep learning Accuracy machine learning algorithms Medical services Feature extraction Classification algorithms History Convolutional neural networks Diseases Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Assessing PM2.5 concentrations in Tehran, Iran, from space using MAIAC, deep blue, and dark target AOD and machine learning algorithms

引用

ATMOSPHERIC POLLUTION RESEARCH 2019年第3期10卷 889-903页

作者： Nabavi, Seyed Omid Haimberger, Leopold Abbasi, Esmail Univ Vienna Dept Meteorol & Geophys Vienna Austria Persian Gulf Univ Dept Environm Studies Bushehr Iran

This study aims to explore the spatial estimation of fine particulate matter (PM2.5) using 10-km merged dark target and deep blue (DB_DT) Aerosol Optical Depth (AOD) and 1-km Multi-Angle Implementation of Atmospheric Correction (MAIAC) AOD over Tehran. The ability of four machine learning algorithms (MLAs) to predict PM2.5 concentrations is also investigated. Results show that the association of satellite AOD with surface PM significantly increases after considering the contribution of relative humidity in PM mass concentration and normalization of AOD to Planetary boundary layer height (PBLH). The examination of derived aerosol layer height (ALH) from 159 Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) profiles shows that PBLH could successfully represent the top of aerosol-laden layers. Surprisingly, the highest correlation was found between normalized 10-km DB_DT AOD and corrected PM2.5 measurements. Consequently, random forest (RF) fed by this AOD product has yielded the best performance (R-2 = 0.68, RMSE = 17.52 and MRE = 27.46%). Importance analysis of variables reveals that DB_DT and meteorological fields are of highest and least importance among selected variables, respectively. The RF performance is less satisfactory during summer which is assumed to be caused by the omission of unknown features representing the formation of secondary aerosols. The inferior accuracy of estimation in the north and east of Tehran is also linked to lacking features which could feed the transportation of PM(2.5 )from west to the east of the study area into MLAs.

关键词： MAIAC AOD DB_DT AOD Vertical distribution of aerosols Fine particulate matters machine learning algorithms

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：