scholarly journals Performance of Robust Linear Classifier with Multivariate Binary Variables

2015 ◽  
Vol 7 (4) ◽  
pp. 104
Author(s):  
I. Egbo ◽  
M. Egbo ◽  
S. I. Onyeagu
Keyword(s):  
Maximum Likelihood ◽  
Discriminant Function ◽  
Likelihood Function ◽  
Error Rates ◽  
Linear Discriminant Function ◽  
Data Set ◽  
Binary Variables ◽  
Linear Discriminant ◽  
Linear Classifiers ◽  
Fisher’S Linear Discriminant

<p>This paper focuses on the robust classification procedures in two group discriminant analysis with multivariate binary variables. A normal distribution based data set is generated using the R-software statistical analysis system 2.15.3 using Barlett’s approximation to chi-square, the data set was found to be homogenous and was subjected to five linear classifiers namely: maximum likelihood discriminant function, fisher’s linear discriminant function, likelihood ratio function, full multinomial function and nearest neighbour function rule. To judge the performance of these procedures, the apparent error rates for each procedure are obtained for different sample sizes. The results obtained ranked the procedures as follows: fisher’s linear discriminant function, maximum likelihood, full multinomial, likelihood function and nearest neigbour function.</p>

The Performance of the Linear Discriminant Function in Nonoptimal Situations and the Estimation of Classification Error Rates: A Review of Recent Findings

1979 ◽  
Vol 16 (3) ◽  
pp. 370-381 ◽  
Author(s):  
William R. Dillon
Keyword(s):  
Discriminant Function ◽  
Error Rates ◽  
Linear Discriminant Function ◽  
Alternative Methods ◽  
Misclassification Error ◽  
Practical Significance ◽  
Classification Error ◽  
Continuous Variables ◽  
Discrete Variables ◽  
Linear Discriminant

This article is a review of the results, as are available, on the performance of the linear discriminant function in situations where the assumptions of multivariate normality and equal group dispersion structures are violated. Some new results are discussed for the case of classification using discrete variables, and in the case of both binary and continuous variables. In addition, alternative methods which have been proposed, and evaluated, for estimating misclassification error rates are thoroughly reviewed. In all cases, the material is reviewed in terms of practical significance, with particular emphasis on the conditions unfavorable to the performance of each procedure.

scholarly journals Flood susceptibility modeling using Radial Basis Function Classifier and Fisher’s linear discriminant function

2021 ◽  
Author(s):  
Chinh Luu ◽  
Duc Dam Nguyen ◽  
Mahdis Amiri ◽  
Phong Tran Van ◽  
Quynh Duy Bui ◽  
...  
Keyword(s):  
Radial Basis Function ◽  
Discriminant Function ◽  
Basis Function ◽  
Model Development ◽  
Linear Discriminant Function ◽  
Linear Discriminant ◽  
Flood Susceptibility ◽  
Statistical Measures ◽  
Radial Basis ◽  
Fisher’S Linear Discriminant

Floods are among the most frequent highly disastrous hazards affecting life, property, and the environment worldwide. While various models are available to predict flood susceptibility, no model is accurate enough to be used for all flood-prone areas. Model development using different algorithms is a continuous process to improve the prediction accuracy of flood susceptibility. In the study, we used the Radial Basis Function and Fisher’s linear discriminant function to develop a flood susceptibility map for a case study of Quang Binh Province. The model development used ten variables (elevation, slope, curvature, river density, distance from river, geomorphology, land use, flow accumulation, flow direction, and rainfall). For model training and validation, input data was split into a 70:30 ratio according to flood locations. Statistical indexes were used to evaluate model performance such as Receiver Operating Characteristic, the Area Under the ROC Curve, Root Mean Square Error, Accuracy, Sensitivity, Specificity, and Kappa index. Results indicated that the radial basis function classifier model had better performance in predicting flood susceptible areas based on the statistical measures (PPV = 92.00%, NPV = 87.00%, SST = 87.62%, SPF = 91.58%, ACC = 89.50%, Kappa = 0.790, MAE = 0.204, RMSE = 0.292 and AUC = 0.957. Therefore, the radial basis function classifier algorithm model is appropriate for predicting flood susceptibility in Quang Binh Province.

Sign in / Sign up

Export Citation Format

Share Document