Paper Info
Title
Neural Networks vs Logistic Regression: a Comparative Study on a Large Data Set
Abstract
Neural networks and logistic regression have been among the most widely used AI technique in applications of pattern classification.Much has been discussed about if there is any significant difference in between them but much less has been actually done with real-world applications data (large scale) to help settle this matter, with a few exceptions.This paper presents a performance comparison between these two techniques on the market application of credit risk assessment, making use of a large database from an outstanding credit bureau and financial institution (a sample of 180,000 examples).The comparison was carried out through a 30-fold stratified cross-validation process to define the confidence intervals for the performance evaluation. Several metrics were applied both on the optimal decision point and along the continuous output domain.The statistical tests showed that multilayer perceptrons perform better than logistic regression at 95% confidence level, for all the metrics used.
Year
DOI
Venue
2004
10.1109/ICPR.2004.616
ICPR (3)
Keywords
Field
DocType
confidence level,large database,real-world applications data,large data set,comparative study,credit risk assessment,confidence interval,logistic regression,performance evaluation,performance comparison,neural networks vs logistic,large scale,outstanding credit bureau,neural network
Optimal decision,Financial institution,Multinomial logistic regression,Computer science,Artificial intelligence,Artificial neural network,Confidence interval,Statistics,Logistic regression,Perceptron,Statistical hypothesis testing,Machine learning
Conference
ISSN
ISBN
Citations 
1051-4651
0-7695-2128-2
3
PageRank 
References 
Authors
0.42
3
6
Name
Order
Citations
PageRank
Paulo J. L. Adeodato110314.27
Germano C. Vasconcelos210412.25
Adrian L. Arnaud3213.67
Roberto A. F. Santos430.42
Rodrigo C. L. V. Cunha5212.90
Domingos S. M. P. Monteiro6171.73