scholarly journals Using Linear Discriminant Analysis to Classify Snowfall Situations into Avalanching and Non-Avalanching Ones

1977 ◽  
Vol 19 (81) ◽  
pp. 679-680
Author(s):  
N.F. Drozdovskaya
Keyword(s):  
Discriminant Analysis ◽  
Statistical Methods ◽  
Training Sample ◽  
Dispersion Analysis ◽  
Snow Accumulation ◽  
Discriminant Functions ◽  
Prognostic Information ◽  
Individual Parameter ◽  
Linear Discriminant ◽  
Avalanche Forecasting

Abstract The existing methods of predicting avalanche danger often do not meet users’ demands because of the empiric character of the insufficient volume of information used. In such forecasts the contribution of each individual parameter into the prognostic information is unknown, and this is very important when studying such an event as avalanche formation, which is conditioned by a complex interaction of numerous factors, including snow accumulation, the state of snow thickness, and the conditions of its development. It is obvious that such problems can be successfully solved by statistical methods, and that explains the growing interest in numerical methods of avalanche forecasting. Problems of multi-dimensional observations arises in many scientific fields. The method suited for this problem is discriminant analysis, the purpose of which is to divide a multi-dimensional observation vector into predetermined classes. This study considers the prognostic (diagnostic) problems of fresh-snow avalanches released during snowfall or in the two days after it has ceased. The theoretical basis is a complex of statistical methods: correlation and dispersion analysis, “sifting" for the choice of predictors’ informative groups, construction of linear parametric discriminant functions, predictions based on training sample, and verification of discriminant functions based on independent material. The archive used in the study consisted of 500 avalanching cases and 1 300 non-avalanching ones. All situations were grouped according to geomorphological characteristics. Each situation is described by eight meteorological characteristics. The results of classification of snowfall situations into avalanching and non-avalanching ones are as follows: reliability of ρ is from 75% to 91%, H from 0.15 to 0.51; based on independent material the reliability of ρ is from 63% to 85%, H from 0.10 to 0.56. This paper has been accepted in revised form for publication in a later issue of the Journal of Glaciology.

scholarly journals Using Linear Discriminant Analysis to Classify Snowfall Situations into Avalanching and Non-Avalanching Ones

1979 ◽  
Vol 22 (86) ◽  
pp. 127-133
Author(s):  
N.F. Drozdovskaya
Keyword(s):  
Discriminant Analysis ◽  
Statistical Methods ◽  
Training Sample ◽  
Dispersion Analysis ◽  
Snow Accumulation ◽  
Discriminant Functions ◽  
Prognostic Information ◽  
Individual Parameter ◽  
Linear Discriminant ◽  
Avalanche Forecasting

AbstractThe existing methods of predicting avalanche danger often do not meet users’ demands because of the empiric character of the insufficient volume of information used. In such forecasts the contribution of each individual parameter into the prognostic information is unknown, and this is very important when studying such an event as avalanche formation, which is conditioned by a complex interaction of numerous factors, including snow accumulation, snow thickness, and the conditions of its development. It is obvious that such problems can be successfully solved by statistical methods, and that explains the growing interest in numerical methods of avalanche forecasting. Problems of multi-dimensional observations arises in many scientific fields. The method suited for this problem is discriminant analysis, the purpose of which is to divide a multi-dimensional observation vector into predetermined classes.This study considers the prognostic (diagnostic) problems of fresh-snow avalanches released during snowfall or in the two days after it has ceased. The theoretical basis is a complex of statistical methods: correlation and dispersion analysis, “sifting” for the choice of predictors’ informative groups, construction of linear parametric discriminant functions, predictions based on training sample, and verification of discriminant functions based on independent material.The archive used in the study consisted of 500 avalanching cases and 1 300 non-avalanching ones. All situations were grouped according to geomorphological characteristics. Each situation is described by eight meteorological characteristics. The results of classification of snowfall situations into avalanching and non-avalanching ones are as follows: reliability of p is from 75% to 91%, H from 0.15 to 0.51; based on independent material the reliability of p is from 63% to 85%, H from 0. 10 to 0.56.

Using Linear Discriminant Analysis to Classify Snowfall Situations into Avalanching and Non-Avalanching Ones

1979 ◽  
Vol 22 (86) ◽  
pp. 127-133
Author(s):  
N.F. Drozdovskaya
Keyword(s):  
Discriminant Analysis ◽  
Statistical Methods ◽  
Training Sample ◽  
Dispersion Analysis ◽  
Snow Accumulation ◽  
Discriminant Functions ◽  
Prognostic Information ◽  
Individual Parameter ◽  
Linear Discriminant ◽  
Avalanche Forecasting

AbstractThe existing methods of predicting avalanche danger often do not meet users’ demands because of the empiric character of the insufficient volume of information used. In such forecasts the contribution of each individual parameter into the prognostic information is unknown, and this is very important when studying such an event as avalanche formation, which is conditioned by a complex interaction of numerous factors, including snow accumulation, snow thickness, and the conditions of its development. It is obvious that such problems can be successfully solved by statistical methods, and that explains the growing interest in numerical methods of avalanche forecasting. Problems of multi-dimensional observations arises in many scientific fields. The method suited for this problem is discriminant analysis, the purpose of which is to divide a multi-dimensional observation vector into predetermined classes.This study considers the prognostic (diagnostic) problems of fresh-snow avalanches released during snowfall or in the two days after it has ceased. The theoretical basis is a complex of statistical methods: correlation and dispersion analysis, “sifting” for the choice of predictors’ informative groups, construction of linear parametric discriminant functions, predictions based on training sample, and verification of discriminant functions based on independent material.The archive used in the study consisted of 500 avalanching cases and 1 300 non-avalanching ones. All situations were grouped according to geomorphological characteristics. Each situation is described by eight meteorological characteristics. The results of classification of snowfall situations into avalanching and non-avalanching ones are as follows: reliability ofpis from 75% to 91%,Hfrom 0.15 to 0.51; based on independent material the reliability ofpis from 63% to 85%,Hfrom 0. 10 to 0.56.

scholarly journals Using Linear Discriminant Analysis to Classify Snowfall Situations into Avalanching and Non-Avalanching Ones

1977 ◽  
Vol 19 (81) ◽  
pp. 679-680
Author(s):  
N.F. Drozdovskaya
Keyword(s):  
Discriminant Analysis ◽  
Statistical Methods ◽  
Training Sample ◽  
Dispersion Analysis ◽  
Snow Accumulation ◽  
Discriminant Functions ◽  
Prognostic Information ◽  
Individual Parameter ◽  
Linear Discriminant ◽  
Avalanche Forecasting

AbstractThe existing methods of predicting avalanche danger often do not meet users’ demands because of the empiric character of the insufficient volume of information used. In such forecasts the contribution of each individual parameter into the prognostic information is unknown, and this is very important when studying such an event as avalanche formation, which is conditioned by a complex interaction of numerous factors, including snow accumulation, the state of snow thickness, and the conditions of its development.It is obvious that such problems can be successfully solved by statistical methods, and that explains the growing interest in numerical methods of avalanche forecasting. Problems of multi-dimensional observations arises in many scientific fields. The method suited for this problem is discriminant analysis, the purpose of which is to divide a multi-dimensional observation vector into predetermined classes.This study considers the prognostic (diagnostic) problems of fresh-snow avalanches released during snowfall or in the two days after it has ceased. The theoretical basis is a complex of statistical methods: correlation and dispersion analysis, “sifting" for the choice of predictors’ informative groups, construction of linear parametric discriminant functions, predictions based on training sample, and verification of discriminant functions based on independent material.The archive used in the study consisted of 500 avalanching cases and 1 300 non-avalanching ones. All situations were grouped according to geomorphological characteristics. Each situation is described by eight meteorological characteristics. The results of classification of snowfall situations into avalanching and non-avalanching ones are as follows: reliability of ρ is from 75% to 91%, H from 0.15 to 0.51; based on independent material the reliability of ρ is from 63% to 85%, H from 0.10 to 0.56.This paper has been accepted in revised form for publication in a later issue of the Journal of Glaciology.

Relationships between Hordeum bulbosum L. subsp. bulbosum and H. bulbosum subsp. nodosum comb, et stat. nov.

1985 ◽  
Vol 63 (4) ◽  
pp. 735-743 ◽  
Author(s):  
Bernard R. Baum ◽  
L. Grant Bailey
Keyword(s):  
Data Analysis ◽  
Discriminant Analysis ◽  
Exploratory Data Analysis ◽  
Correct Identification ◽  
Hordeum Bulbosum ◽  
Discriminant Functions ◽  
Morphometric Characters ◽  
Linear Discriminant ◽  
Group Assignment ◽  
Exploratory Data

Material of the diploid (HBD) and tetraploid (HBT) Hordeum bulbosum collected in the Mediterranean and Near East areas was examined for 14 morphometric characters. Exploratory data analysis revealed that cilia on the margins of the glumes of the central spikelets may or may not be present in HBT but are never present in HBD. The data were submitted to various kinds of discriminant analysis in which group assignment was based on ploidy level. When presence–absence of cilia on glume margins is used in combination with the resulting linear discriminant functions (DF) there is about 91% probability of correct identification; with DF alone there is about 81% probability. The results of discriminant analysis provided justification, in the opinion of the authors, to regard HBD and HBT as separate taxa at the level of subspecies, namely H. bulbosum subsp. bulbosum and H. bulbosum L. subsp. nodosum (L.) Baum.

Biochemical individuality and the recognition of personal profiles with a computer.

1980 ◽  
Vol 26 (1) ◽  
pp. 30-36 ◽  
Author(s):  
E A Robertson ◽  
A C Van Steirteghem ◽  
J E Byrkit ◽  
D S Young
Keyword(s):  
Discriminant Analysis ◽  
Continuous Flow ◽  
Nonlinear Mapping ◽  
Biochemical Profile ◽  
Discriminant Functions ◽  
Linear Discriminant ◽  
Testing Period ◽  
Linear Maps ◽  
Non Linear

Abstract Multitest analysis of an individual's blood provides a biochemical profile that reflects his identity and pathophysiological state. During a six-week period we repeatedly profiled 10 volunteers for 22 different analytes, using continuous-flow and discrete analyzers (SMAC, KA 150 enzyme analyzer, ABA-100, AutoAnalyzers) and manual procedures. Two years later, we obtained multiple follow-up profiles. Using linear discriminant functions derived from the first five (or first 10) specimens from each subject, we were able correctly to identify 96% (or 100%) of the specimens collected during the remainder of the six-week testing period. Ninety percent of the two-year follow-up specimens were correctly identified when we used all the original profiles to calculate the discriminant functions. Deliberately mislabeled specimens were also correctly identified by discriminant analysis. Profiles of individual samples (and average profiles for each subject) were graphically displayed as computer-drawn faces and non-linear maps. Covariances between pairs of tests on repeated profiles differed significantly for different subjects. Inter-test relationships were graphically displayed by nonlinear mapping.

Estimation of Error Rates in Several-Population Discriminant Analysis

1982 ◽  
Vol 19 (1) ◽  
pp. 57-61 ◽  
Author(s):  
Stephen C. Hora ◽  
James B. Wilcox
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Discriminant Analysis ◽  
Classification Accuracy ◽  
Posterior Probability ◽  
Error Rates ◽  
Discriminant Functions ◽  
Linear Discriminant ◽  
Probability Estimator ◽  
Estimation Of Error

Researchers seeking to estimate the classification accuracy of linear discriminant functions in a more than two-population setting have had little guidance as to the most appropriate technique. The authors review the available techniques and present an additional alternative which combines features of the U-method and the recently developed posterior probability estimator. The new alternative is compared with other methods by Monté Carlo simulation.

scholarly journals Predicting Mechanical Ventilation and Mortality in COVID-19 Using Radiomics and Deep Learning on Chest Radiographs: A Multi-Institutional Study

Diagnostics ◽  
2021 ◽  
Vol 11 (10) ◽  
pp. 1812
Author(s):  
Joseph Bae ◽  
Saarthak Kapse ◽  
Gagandeep Singh ◽  
Rishabh Gattu ◽  
Syed Ali ◽  
...  
Keyword(s):  
Mechanical Ventilation ◽  
Deep Learning ◽  
Discriminant Analysis ◽  
Medical Center ◽  
Characteristic Curve ◽  
Chest Radiographs ◽  
University Hospital ◽  
Prognostic Information ◽  
Linear Discriminant ◽  
Severity Scores

In this study, we aimed to predict mechanical ventilation requirement and mortality using computational modeling of chest radiographs (CXRs) for coronavirus disease 2019 (COVID-19) patients. This two-center, retrospective study analyzed 530 deidentified CXRs from 515 COVID-19 patients treated at Stony Brook University Hospital and Newark Beth Israel Medical Center between March and August 2020. Linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and random forest (RF) machine learning classifiers to predict mechanical ventilation requirement and mortality were trained and evaluated using radiomic features extracted from patients’ CXRs. Deep learning (DL) approaches were also explored for the clinical outcome prediction task and a novel radiomic embedding framework was introduced. All results are compared against radiologist grading of CXRs (zone-wise expert severity scores). Radiomic classification models had mean area under the receiver operating characteristic curve (mAUCs) of 0.78 ± 0.05 (sensitivity = 0.72 ± 0.07, specificity = 0.72 ± 0.06) and 0.78 ± 0.06 (sensitivity = 0.70 ± 0.09, specificity = 0.73 ± 0.09), compared with expert scores mAUCs of 0.75 ± 0.02 (sensitivity = 0.67 ± 0.08, specificity = 0.69 ± 0.07) and 0.79 ± 0.05 (sensitivity = 0.69 ± 0.08, specificity = 0.76 ± 0.08) for mechanical ventilation requirement and mortality prediction, respectively. Classifiers using both expert severity scores and radiomic features for mechanical ventilation (mAUC = 0.79 ± 0.04, sensitivity = 0.71 ± 0.06, specificity = 0.71 ± 0.08) and mortality (mAUC = 0.83 ± 0.04, sensitivity = 0.79 ± 0.07, specificity = 0.74 ± 0.09) demonstrated improvement over either artificial intelligence or radiologist interpretation alone. Our results also suggest instances in which the inclusion of radiomic features in DL improves model predictions over DL alone. The models proposed in this study and the prognostic information they provide might aid physician decision making and efficient resource allocation during the COVID-19 pandemic.

scholarly journals MATHEMATICAL MODEL OF SELECTION OF PHENOTYPES OF BRONCHIAL ASTHMA AT THE RESORT STAGE OF MEDICAL REHABILITATION

2018 ◽  
Vol 1 (68) ◽  
pp. 17-22
Author(s):  
Лейла Дудченко ◽  
Leila Dudchenko ◽  
Валентин Савченко ◽  
Valentin Savchenko
Keyword(s):  
Mathematical Model ◽  
Discriminant Analysis ◽  
Bronchial Asthma ◽  
Medical Rehabilitation ◽  
Discriminant Functions ◽  
Linear Discriminant ◽  
Arterial Blood ◽  
6 Minute Walk Test ◽  
The Mathematical Model ◽  
Selection Of

The aim of the research is to develop a mathematical model of selection of phenotypes of bronchial asthma (BA) at a resort stage of medical rehabilitation. Materials and methods: 300 patients with BA who arrived at the climatic resort for medical rehabilitation; values of 64 indices of the research describing features of emergence and a course of a disease; the current clinic-functional state of patients; the discriminant analysis for the creation of a mathematical model. Results of the research: by the discriminant analysis there were created 7 statistically significant linear discriminant functions for the discernment of BA phenotypes; 23 indices of the research were under study: gender, asthma control, quantity of not allergic associated diseases, coughing during the day, character of the sputum, expressiveness of symptoms of asthma, use of rescue medication, the number of dry rattles in lungs, severity of hypostases, systolic and diastolic arterial blood pressure, shortness of breath, wheezing, reaction to the change of weather, allergic reactions, existence of symptoms of intoxication and feature of treatment of the last exacerbation, a dose of inhaled glucocorticosteroids, presence of emphysema and fibrosis in lungs at the X-ray examination, expressiveness of electrographic changes, FEF25-75%, 6-minute walk test. Conclusion: the procedure of the selection of 7 BA phenotypes at the resort stage of medical rehabilitation can be carried out by the use of the mathematical model consisting of 7 linear discriminant functions including 23 indices; the accuracy of the selection of the offered BA phenotypes of the developed mathematical model in general is 89.0%.

scholarly journals Soil-Agroecological Types of Lands of the Petrovsky District in the Tambov Region of Russia

2021 ◽  
Author(s):  
Anna Murman
Keyword(s):  
Discriminant Analysis ◽  
Linear Discriminant Analysis ◽  
Soil Cover ◽  
Functional Organization ◽  
Landscape Planning ◽  
Training Sample ◽  
Forest Steppe ◽  
Moisture Regime ◽  
Simulation Methods ◽  
Linear Discriminant

This article describes methodological approaches and results for digital mapping of water-migrational and erosional-accumulative soil cover structures for the forest-steppe of Tambov Plain. Such maps form the basis for applied maps such as for agroecological studies, forestry, landscape planning, etc. In this study, soil-landscape relationships were simulated as one of the subsystems of structure-functional organization. Linear discriminant analysis, random forest and the supported vector machine were used as simulation methods. The training sample consisted of 256 soil points. The Digital Elevation Model (DEM) had a spatial resolution of 25×25 meters. The simulation was provided for interfluves and valleys separately. A number of factors that describe soil cover type formation within interfluves and valleys were determined. It was established that within interfluves, determinant covariates are linked with moisture regime, whereas factors of lateral transfer and accumulation are most significant within valleys. The hierarchical nature of structure-functional organization was determined. The comparison of the results of the three simulation methods showed that the supported vector machine had the best accuracy values. However, verification by soil maps had the best correlations with the results of the linear discriminant analysis. In addition, soil-agroecological types of lands and their detailed descriptions for the key area were proposed on the basis of the simulation results of the soil combinations. Keywords: soil-agroecological types, soil cover structures, landscape-adaptive agriculture, digital soil mapping

Sign in / Sign up

Export Citation Format

Share Document