Comparative study on the machine learning approaches for the prognosis of acute inflammations in urinary bladder

The incidence of urinary bladder inflammation in recent years is emerging due to the presence of various bacteria. Therefore it is necessary to develop a user interface model for the physicians to make automated prognostic decisions depending on the questionnaire based variables. This prognosis may assist on the diagnostic decision. Hence, in this work we estimated the performance of 3 different classification algorithms in the available dataset that describes physiological and healthy and bladder inflammation. The algorithms such as Support Vector Machines (SVM), Logistic Regression (LR) and Naïve Bayes (NB) exhibited an accuracy of 100% with all similar performance metrics. However there is a difference in the errors metrics with NB showing higher error compared to other two classification algorithms. These results highlighted the dependency of automated decision making and can be used in clinical setup after the establishment of user interface.