A Normalized Probabilistic Expectation-Maximization Neural Network for Minimizing Bayesian Misclassification Cost Risk

Authors:
Parag C. Pendharkar
Affiliations:
School of Business Administration, Pennsylvania State University at Harrisburg, Middletown, USA 17057
Venue:
Neural Processing Letters
Year:
2013

Citing 8
Cited 0

Probabilistic neural networks

Neural Networks
Computer systems that learn: classification and prediction methods from statistics, neural nets, machine learning, and expert systems

Computer systems that learn: classification and prediction methods from statistics, neural nets, machine learning, and expert systems
Mathematical Programming for Data Mining: Formulations and Challenges

INFORMS Journal on Computing
Support Vector Machine Soft Margin Classifiers: Error Analysis

The Journal of Machine Learning Research
A mixed integer optimisation model for data classification

Computers and Industrial Engineering
Data Mining: Practical Machine Learning Tools and Techniques

Data Mining: Practical Machine Learning Tools and Techniques
Data Mining: Concepts and Techniques

Data Mining: Concepts and Techniques
Probabilistic Approaches For Credit Screening And Bankruptcy Prediction

International Journal of Intelligent Systems in Accounting and Finance Management

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a normalized semi-supervised probabilistic expectation-maximization neural network (PEMNN) that minimizes Bayesian misclassification cost risk. Using simulated and real-world datasets, we compare the proposed PEMNN with supervised cost sensitive probabilistic neural network (PNN), discriminant analysis (DA), mathematical integer programming (MIP) model and support vector machines (SVM) for different misclassification cost asymmetries and class biases. The results of our experiments indicate that the PEMNN performs better when class data distributions are normal or uniform. However, when class data distribution is exponential the performance of PEMNN deteriorates giving slight advantage to competing MIP, DA, PNN and SVM techniques. For real-world data with non-parametric distributions and mixed decision-making attributes (continuous and categorical), the PEMNN outperforms the PNN.