Application of Bayesian logistic regression to mining biomedical data

Avali VR, Cooper GF, Gopalakrishnan V. Application of Bayesian logistic regression to mining biomedical data. In: Proceedings of the Fall Symposium of the American Medical Informatics Association (2014) Nov 14;2014:266-73. PMID: 25954328 PMC4419893

Mining high dimensional biomedical data with existing classifiers is challenging and the predictions are often inaccurate. We investigated the use of Bayesian Logistic Regression (B-LR) for mining such data to predict and classify various disease conditions. The analysis was done on twelve biomedical datasets with binary class variables and the performance of B-LR was compared to those from other popular classifiers on these datasets with 10-fold cross validation using the WEKA data mining toolkit.  The statistical significance of the results was analyzed by paired two tailed t-tests and non-parametric Wilcoxon signed-rank tests. We observed overall that B-LR with noninformative Gaussian priors performed on par with other classifiers in terms of accuracy, balanced accuracy and AUC. These results suggest that it is worthwhile to explore the application of B-LR to predictive modeling tasks in bioinformatics using informative biological prior probabilities. With informative prior probabilities, we conjecture that the performance of B-LR will improve.

Publication Year: 
Publication Credits: 
Avali VR, Cooper GF, Gopalakrishnan V.
PDF icon Avali.pdf308.5 KB