A Hybrid Ensemble Feature Selection-Based Learning Model for COPD Prediction on High-Dimensional Feature Space

Chronic obstructive pulmonary disease (COPD) can be identified by airflow limitation in lungs, and this condition is not at all reversible. Generally, the diagnosis process of COPD completely depends upon symptoms, medical history, clinical examination, and lung airflow obstruction. Hence, it is essential to develop an appropriate and effective automatic test in order to diagnose COPD for better disease management. Most of the traditional auto-classification models such as SVM, neural network, Bayesian model, expectation-maximization, and random tree are used to find the appropriate features and its decision patterns for COPD detection. However, as the number of COPD features increases, these models require high-computational memory and time for feature selection and pattern evaluation. Also these models generate high false-positive rate and error rate due to high feature space and data uncertainty. In order to overcome these issues, a hybrid ensemble feature selection-based classification model is proposed on high-dimensional dataset. Experimental results proved that the present model improves the true positivity and error rate compared to the traditional models.