Attribute Selection Model using Meta Classifiers for Pneumonia Diagnosis System

Pneumonia which is caused by severe acute respiratory syndrome. It is also one of the worst pandemics in recent history. Identification of pneumonia affected patients or suspected patients is becoming very crucial nowadays. The proposed article is to trained and test a machine learning model to detect pneumonia based on chest X-ray images. The model built using machine learning algorithms like Meta classifiers are trained on the data sets containing more than 5863 chest X-ray images composed of two categories as pneumonia affected and normal patients. X ray images are the anterior and posterior images that are selected from the longitudinal reviews from patients of 1 to 5 years old from Guangzhou women and children’s Medical Centre. The test x-ray performed as the part of patience routine clinical care during this covid-19. The Proposed system of Pneumonia detection is done with and without attribute selection methodology. A clear analytical system is formed by building a model using some predefined Meta classifiers using machine learning algorithms. Algorithms used are Bagging, AdaBoost, LogitBoost, and Random Committee and Random Subspace algorithms. The accuracy of the built model obtained during the classification process is 89.27% for bagging algorithm and the receiver operator characteristics for all the above mention classifiers is greater than 0.9. The performance of the selected Meta classifiers accuracy level of the classification after selecting the attributes lies from 84.819% to 88.9354% for Ada Boost and Random subspace classifiers.