Implementation of Multivariate Exponential Power Distribution in Discrimination and Classification of Psychological Data and Other applications

Recent advances have shown that some multivariate psychological data are deviating from usual normal assumption either in the tails or kurtosis. Thereby, allowing the call for modelling of such data using more robust elliptically contoured density which includes the normal distribution as a special case. This allowed more flexibility at the kurtosis and tail regions, which is better in handling non-normality in data analysis and also lower the cost of misclassification. The present study employed a robust model for such cases in the context of discrimination and classification of multivariate psychological disorder data using multivariate exponential distribution as an underlining model. Parameters were estimated using the method of maximum likelihood estimation and the discrimination and classification were based on the log likelihood ratio approach. The resulting models relied solidly on the shape parameter, which regulate the tails and the kurtosis, thereby allowed flexibility. This method enable us to lower the cost of misclassification. Some other areas of applications were also considered in the paper.

Adaptive covariate acquisition for minimizing total cost of classification

In some applications, acquiring covariates comes at a cost which is not negligible. For example in the medical domain, in order to classify whether a patient has diabetes or not, measuring glucose tolerance can be expensive. Assuming that the cost of each covariate, and the cost of misclassification can be specified by the user, our goal is to minimize the (expected) total cost of classification, i.e. the cost of misclassification plus the cost of the acquired covariates. We formalize this optimization goal using the (conditional) Bayes risk and describe the optimal solution using a recursive procedure. Since the procedure is computationally infeasible, we consequently introduce two assumptions: (1) the optimal classifier can be represented by a generalized additive model, (2) the optimal sets of covariates are limited to a sequence of sets of increasing size. We show that under these two assumptions, a computationally efficient solution exists. Furthermore, on several medical datasets, we show that the proposed method achieves in most situations the lowest total costs when compared to various previous methods. Finally, we weaken the requirement on the user to specify all misclassification costs by allowing the user to specify the minimally acceptable recall (target recall). Our experiments confirm that the proposed method achieves the target recall while minimizing the false discovery rate and the covariate acquisition costs better than previous methods.

Feature selected cost-sensitive twin SVM for imbalanced data

In this paper, we propose a cost-sensitive twin SVM (cs-tsvm) and apply it to imbalanced data. A weight is added to each instance according to its cost of misclassification which is related to its position. In preprocessing part, features are selected by their difference of majority and minority classes. The feature is selected when its difference value is higher than average one. The experiment is conducted on UCI datasets and G-mean, AUC and accuracy are evaluation metrics. The experimental results show that Feature selection with CS-TWSVM is useful for datasets with high dimension.

Screening Toxic Chemicals: How Accurate Must Tests Be?

A decision analysis framework is used to explore the value of screening tests for carcinogenicity. Whether a test lowers the social cost of screening depends on the test's sensitivity, specificity, and cost and the social cost of misclassification (exonerating carcinogenic chemicals or condemning noncarcinogenic chemicals). The model shows that the best screening test need not be either the most accurate or the least expensive.

Pattern Recognition Methods for Optimizing Multivariate Tissue Signatures in Diagnostic Ultrasound

Described is a supervised parametric approach to the detection and classification of disease from acoustic data. Statistical pattern recognition techniques are implemented to design the best ultrasonic tissue signature from a set of measurements and for a given task, and to rate its performance in a way that can be compared with other diagnostic tools. In this paper, we considered combinations of four ultrasonic tissue parameters to discriminate, in vivo, between normal liver and chronic active hepatitis. The separation between normal and diseased samples was made by application of the Bayes decision rule for minimum risk which includes the prior probability for the presence of disease and the cost of misclassification. Large differences in classification performance of various tissue parameter combinations were demonstrated using the Hotelling trace criterion (HTC) and receiver operating characteristic (ROC) analysis. The ability of additional measurements to increase or decrease discriminability, even measurements from other diagnostic modalities, can be evaluated directly in this manner.