Analyzing the balancing of error rates for multi-group classification

Authors:
Dinesh R. Pai;Kenneth D. Lawrence;Ronald K. Klimberg;Sheila M. Lawrence
Affiliations:
School of Business Administration, Pennsylvania State University at Harrisburg, 777 West Harrisburg Pike, Middletown, PA 17057, USA;School of Management, New Jersey Institute of Technology, Newark, NJ 07102, USA;Haub School of Business, Saint Joseph's University, Philadelphia, PA 19066, USA;MSIS Department, Rutgers University, Piscataway, NJ 08854, USA
Venue:
Expert Systems with Applications: An International Journal
Year:
2012

Citing 9
Cited 0

Managerial applications of neural networks: the case of bank failure predictions

Management Science
Computational characteristics of a new mathematical programming model for the three-group discriminant problem

Computers and Operations Research
An Application of Linear Programming Discriminant Analysis to Classifying and Predicting the Symptomatic Status of HIV/AIDS Patients

Journal of Medical Systems
A simple weighting scheme for classification in two-group discriminant problems

Computers and Operations Research
A comparative assessment of classification methods

Decision Support Systems
Predicting the type of pregnancy using artificial neural networks and multinomial logistic regression: a comparison study

Neural Computing and Applications
Review: Neural networks and statistical techniques: A review of applications

Expert Systems with Applications: An International Journal
A new mathematical programming approach to multi-group classification problems

Computers and Operations Research
Experimental comparison of parametric, non-parametric, and hybrid multigroup classification

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	12.05

Visualization

Abstract

This paper reports the relative performance of an experimental comparison of some well-known classification techniques such as classical statistical, artificial intelligence, mathematical programming (MP), and hybrid approaches. In particular, we examine the four-group, three-variable problem and the associated error rates for the four groups when each of the models is applied to various sets of simulated data. The data had varying characteristics such as multicollinearity, nonlinearity, sample proportions, etc. We concentrate on individual error rates for the four groups, i.e., we count the number of group 1 values classified into group 2, group 3, and group 4 and vice versa. The results indicate that in general not only are MP, k-NN, and hybrid approaches relatively better at overall classification but they also provide a much better balance between error rates for the top customer groups. The results also indicate that the MP and hybrid approaches provide relatively higher and stable classification accuracy under all the data characteristics.