A Machine Learning Approach to Modeling Tropical Cyclone Wind Field Uncertainty

2017 ◽  
Vol 145 (8) ◽  
pp. 3203-3221 ◽  
Author(s):  
Thomas Loridan ◽  
Ryan P. Crompton ◽  
Eugene Dubossarsky
Keyword(s):  
Machine Learning ◽  
Risk Assessment ◽  
Tropical Cyclone ◽  
Wind Field ◽  
Full Range ◽  
Wrf Model ◽  
Principal Component ◽  
Machine Learning Algorithms ◽  
Probabilistic Forecasting ◽  
Wind Fields

Tropical cyclone (TC) risk assessment models and probabilistic forecasting systems rely on large ensembles to simulate the track trajectories, intensities, and spatial distributions of damaging winds from severe events. Given computational constraints associated with the generation of such ensembles, the representation of TC winds is typically based on very simple parametric formulations. Such models strongly underestimate the full range of TC wind field variability and thus do not allow for accurate representation of the risk profile. With this in mind, this study explores the potential of machine learning algorithms as an alternative to current parametric methods. First, a catalog of high-resolution TC wind simulations is assembled for the western North Pacific using the Weather Research and Forecasting (WRF) Model. The simulated wind fields are then decomposed via principal component analysis (PCA) and a quantile regression forest model is trained to predict the conditional distributions of the first three principal component (PC) weights. With this model, predictions can be made for any quantiles in the distributions of the PC weights thereby providing a way to account for uncertainty in the modeled wind fields. By repeatedly sampling the quantile values, probabilistic maps for the likelihood of attaining given wind speed thresholds can be easily generated. Similarly the inclusion of such a model as part of a TC risk assessment framework can greatly increase the range of wind field patterns sampled, providing a broader view of the threat posed by TC winds.

Implementing machine learning algorithms for suicide risk prediction in clinical practice: A focus group study

2021 ◽  
Author(s):  
Kate Bentley ◽  
Kelly Zuromski ◽  
Rebecca Fortgang ◽  
Emily Madsen ◽  
Daniel Kessler ◽  
...  
Keyword(s):  
Machine Learning ◽  
Risk Assessment ◽  
Clinical Practice ◽  
Risk Prediction ◽  
Suicide Risk ◽  
Prediction Models ◽  
Clinical Decision ◽  
Machine Learning Algorithms ◽  
Suicide Risk Assessment ◽  
Focus Group Study

Background: Interest in developing machine learning algorithms that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. Whether and how such models might be implemented and useful in clinical practice, however, remains unknown. In order to ultimately make automated suicide risk prediction algorithms useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders (including the frontline providers who will be using such tools) at each stage of the implementation process.Objective: The aim of this focus group study was to inform ongoing and future efforts to deploy suicide risk prediction models in clinical practice. The specific goals were to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk prediction algorithms; and identify barriers, facilitators, recommendations, and factors to consider for initiatives in this area. Methods: We conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by two independent study staff members. All coded text was reviewed and discrepancies resolved in consensus meetings with doctoral-level staff. Results: Though most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the healthcare system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider trainings. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings.Conclusions: Providers were dissatisfied with current suicide risk assessment methods and open to the use of a machine learning-based risk prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of new methods in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.

Hybridization of Machine Learning Algorithm in Intrusion Detection System

2020 ◽  
pp. 150-175
Author(s):  
Amudha P. ◽  
Sivakumari S.
Keyword(s):  
Machine Learning ◽  
Intrusion Detection ◽  
Classification Accuracy ◽  
Intrusion Detection System ◽  
Learning Algorithm ◽  
Detection System ◽  
Principal Component ◽  
Machine Learning Algorithms ◽  
Feature Selection Technique ◽  
Efficient Manner

In recent years, the field of machine learning grows very fast both on the development of techniques and its application in intrusion detection. The computational complexity of the machine learning algorithms increases rapidly as the number of features in the datasets increases. By choosing the significant features, the number of features in the dataset can be reduced, which is critical to progress the classification accuracy and speed of algorithms. Also, achieving high accuracy and detection rate and lowering false alarm rates are the major challenges in designing an intrusion detection system. The major motivation of this work is to address these issues by hybridizing machine learning and swarm intelligence algorithms for enhancing the performance of intrusion detection system. It also emphasizes applying principal component analysis as feature selection technique on intrusion detection dataset for identifying the most suitable feature subsets which may provide high-quality results in a fast and efficient manner.

scholarly journals Evaluating Variable Selection and Machine Learning Algorithms for Estimating Forest Heights by Combining Lidar and Hyperspectral Data

2020 ◽  
Vol 9 (9) ◽  
pp. 507
Author(s):  
Sanjiwana Arjasakusuma ◽  
Sandiaga Swahyu Kusuma ◽  
Stuart Phinn
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Learning Algorithms ◽  
Principal Component ◽  
Hyperspectral Data ◽  
Machine Learning Algorithms ◽  
Gradient Boosting ◽  
Support Vector ◽  
Forest Height ◽  
Extreme Gradient Boosting

Machine learning has been employed for various mapping and modeling tasks using input variables from different sources of remote sensing data. For feature selection involving high- spatial and spectral dimensionality data, various methods have been developed and incorporated into the machine learning framework to ensure an efficient and optimal computational process. This research aims to assess the accuracy of various feature selection and machine learning methods for estimating forest height using AISA (airborne imaging spectrometer for applications) hyperspectral bands (479 bands) and airborne light detection and ranging (lidar) height metrics (36 metrics), alone and combined. Feature selection and dimensionality reduction using Boruta (BO), principal component analysis (PCA), simulated annealing (SA), and genetic algorithm (GA) in combination with machine learning algorithms such as multivariate adaptive regression spline (MARS), extra trees (ET), support vector regression (SVR) with radial basis function, and extreme gradient boosting (XGB) with trees (XGbtree and XGBdart) and linear (XGBlin) classifiers were evaluated. The results demonstrated that the combinations of BO-XGBdart and BO-SVR delivered the best model performance for estimating tropical forest height by combining lidar and hyperspectral data, with R2 = 0.53 and RMSE = 1.7 m (18.4% of nRMSE and 0.046 m of bias) for BO-XGBdart and R2 = 0.51 and RMSE = 1.8 m (15.8% of nRMSE and −0.244 m of bias) for BO-SVR. Our study also demonstrated the effectiveness of BO for variables selection; it could reduce 95% of the data to select the 29 most important variables from the initial 516 variables from lidar metrics and hyperspectral data.

Parametric Modeling of Transitioning Cyclone Wind Fields for Risk Assessment Studies in the Western North Pacific

2015 ◽  
Vol 54 (3) ◽  
pp. 624-642 ◽  
Author(s):  
T. Loridan ◽  
S. Khare ◽  
E. Scherer ◽  
M. Dixon ◽  
E. Bellone
Keyword(s):  
Risk Assessment ◽  
North Pacific ◽  
Wind Field ◽  
Western North Pacific ◽  
Parametric Modeling ◽  
Transition Phase ◽  
Wind Fields ◽  
Extratropical Transition ◽  
Starting Point ◽  
Assessment Systems

AbstractProbabilistic risk assessment systems for tropical cyclone hazards rely on large ensembles of model simulations to characterize cyclones tracks, intensities, and the extent of the associated damaging winds. Given the computational costs, the wind field is often modeled using parametric formulations that make assumptions that are based on observations of tropical systems (e.g., satellite, or aircraft reconnaissance). In particular, for the Northern Hemisphere, most of the damaging contribution is assumed to be from the right of the moving cyclone, with the left-hand-side winds being much weaker because of the direction of storm motion. Recent studies have highlighted that this asymmetry assumption does not hold for cyclones undergoing extratropical transitions around Japan. Transitioning systems can exhibit damaging winds on both sides of the moving cyclone, with wind fields often characterized as resembling a horseshoe. This study develops a new parametric formulation of the extratropical transition phase for application in risk assessment systems. A compromise is sought between the need to characterize the horseshoe shape while keeping the formulation simple to allow for implementation within a risk assessment framework. For that purpose the tropical wind model developed by Willoughby et al. is selected as a starting point and parametric bias correction fields are applied to build the target shape. Model calibration is performed against a set of 37 extratropical transition cases simulated using the Weather Research and Forecasting Model. This newly developed parametric model of the extratropical transition phase shows an ability to reproduce wind field features observed in the western North Pacific Ocean while using only a restricted number of input parameters.

Improved Tropical-Cyclone Flight-Level Wind Estimates Using Routine Infrared Satellite Reconnaissance

2015 ◽  
Vol 54 (2) ◽  
pp. 463-478 ◽  
Author(s):  
John A. Knaff ◽  
Scott P. Longmore ◽  
Robert T. DeMaria ◽  
Debra A. Molenar
Keyword(s):  
Tropical Cyclone ◽  
Principal Components ◽  
Wind Field ◽  
Motion Vector ◽  
Principal Component ◽  
Principal Component Method ◽  
Tangential Wind ◽  
Flight Level ◽  
Polar Grid ◽  
Single Field

AbstractA new and improved method for estimating tropical-cyclone (TC) flight-level winds using globally and routinely available TC information and infrared (IR) satellite imagery is presented. The developmental dataset is composed of aircraft reconnaissance (1995–2012) that has been analyzed to a 1 km × 10° polar grid that extends outward 165 km from the TC center. The additional use of an azimuthally average tangential wind at 500 km, based on global model analyses, allows the estimation of winds at larger radii. Analyses are rotated to a direction-relative framework, normalized by dividing the wind field by the observed maximum, and then decomposed into azimuthal wavenumbers in terms of amplitudes and phases. Using a single-field principal component method, the amplitudes and phases of the wind field are then statistically related to principal components of motion-relative IR images and factors related to the climatological radius of maximum winds. The IR principal components allow the wind field to be related to the radial and azimuthal variability of the wind field. Results show that this method, when provided with the storm location, the estimated TC intensity, the TC motion vector, and a single IR image, is able to estimate the azimuthal wavenumber 0 and 1 components of the wind field. The resulting wind field reconstruction significantly improves on the method currently used for satellite-based operational TC wind field estimates. This application has several potential uses that are discussed within.

Cyclone Wind Field Asymmetries during Extratropical Transition in the Western North Pacific

2014 ◽  
Vol 53 (2) ◽  
pp. 421-428 ◽  
Author(s):  
T. Loridan ◽  
E. Scherer ◽  
M. Dixon ◽  
E. Bellone ◽  
S. Khare
Keyword(s):  
Risk Assessment ◽  
North Pacific ◽  
Wind Field ◽  
Western North Pacific ◽  
North Pacific Ocean ◽  
Maximum Magnitude ◽  
Wind Fields ◽  
Extratropical Transition ◽  
Left Hand ◽  
Assessment Systems

AbstractRisk-assessment systems for wind hazards (e.g., hurricanes or typhoons) often rely on simple parametric wind field formulations. They are built using extensive observations of tropical cyclones and make assumptions about wind field asymmetry. In this framework, maximum winds are always simulated to the right of the cyclone, but analysis of the Climate Forecast System Reanalysis database for the western North Pacific Ocean suggests that wind fields from cyclones undergoing extratropical transition around Japan often present features that cannot be adequately simulated under these assumptions. These “left-hand-side contribution” (LHSC) wind fields exhibit strong winds on both sides of the moving cyclone with the maximum magnitude often located to the left. Classification of cyclones in terms of their most frequent patterns reveals that 67% of cases that make a transition around Japan are dominantly LHSC. They are more likely in autumn and have more intense maximum winds. The results from this study show the need for a new approach to the modeling of transitioning wind fields in the context of risk-assessment systems.

scholarly journals Predicting Forest Fires using Supervised and Ensemble Machine Learning Algorithms

2019 ◽  
Vol 8 (2) ◽  
pp. 3697-3705 ◽  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Forest Fires ◽  
Principal Component ◽  
Climatic Conditions ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Gradient Boosting ◽  
Support Vector ◽  
Physical Factors

Forest fires have become one of the most frequently occurring disasters in recent years. The effects of forest fires have a lasting impact on the environment as it lead to deforestation and global warming, which is also one of its major cause of occurrence. Forest fires are dealt by collecting the satellite images of forest and if there is any emergency caused by the fires then the authorities are notified to mitigate its effects. By the time the authorities get to know about it, the fires would have already caused a lot of damage. Data mining and machine learning techniques can provide an efficient prevention approach where data associated with forests can be used for predicting the eventuality of forest fires. This paper uses the dataset present in the UCI machine learning repository which consists of physical factors and climatic conditions of the Montesinho park situated in Portugal. Various algorithms like Logistic regression, Support Vector Machine, Random forest, K-Nearest neighbors in addition to Bagging and Boosting predictors are used, both with and without Principal Component Analysis (PCA). Among the models in which PCA was applied, Logistic Regression gave the highest F-1 score of 68.26 and among the models where PCA was absent, Gradient boosting gave the highest score of 68.36.

scholarly journals EKMPRFG: Ensemble of KNN, Multilayer Perceptron and Random Forest using Grading for Android Malware Classification

2020 ◽  
Vol 8 (5) ◽  
pp. 3353-3360
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Standard Deviation ◽  
Principal Component ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Android Malware ◽  
Android Malware Detection ◽  
Significant Research

Android is the most popular Operating Systems with over 2.5 billion devices across the globe. The popularity of this OS has unfortunately made the devices and the services they enable, vulnerable to numerous security threats. As a result of this, a significant research is being done in the field of Android Malware Detection employing Machine Learning Algorithms. Our current work emphasizes on the possible use of Machine Learning techniques for the detection of malware on such android devices. The proposed EKMPRFG is applied for the classification of Android Malware after a preprocessing phase involving a hybrid Feature Selection model using proposed Standard Deviation of Standard Deviation of Ranks (SDSDR) and several other builtin Feature Selection algorithms such as Correlation based Feature Selection (CFS), Classifier SubsetEval, Consistency SubsetEval, and Filtered SubsetEval followed by Principal Component Analysis(PCA) for dimensionality reduction. The experimental results obtained on two data sets indicate that EKMPRFG outperforms the existing works in terms of Prediction Accuracy and Weighted F- Measure values.

Sign in / Sign up

Export Citation Format

Share Document