Enhancing the accuracy in predicting infant mortality using random forest in comparison with logistic regression

To achieve a higher level of precision in the prediction of infant mortality, we contrasted Random Forest with Logistic Regression methodology. Components and Methods of Research: There are two categories that are included in the study, and they are random forest and logistic regression. The number of observations (N=10) is applied to a dataset that has 750 records, 22805 rows, and 20 columns in order to improve the accuracy of the prediction of the infant mortality rate. The considerable mortality risk that is associated with premature delivery continues to be a relevant worry even in this modern era. There is currently only a small number of static characteristics that are used to calculate clinical survival prediction scores. These variables include gestational age, low birth weight, temperature, and entry hypertension. In addition, as opposed to the models, preterm newborns who are brought to the neonatal intensive care unit (NICU) have access to numerical and minute vital sign data, which may provide even more insight into the outcomes. It is possible for real-time computational models to predict the probability of preterm delivery in the neonatal intensive care unit (NICU) by combining data on vital signs with static clinical parameters. This type of prediction may be more accurate and clinically relevant than static prediction models. At a success rate of 86.53 percent, Novel Random Forest greatly exceeds Logistic Regression in terms of object identification. Logistic Regression has an accuracy rate of 85.29 percent. On the basis of the independent sample t-test, both techniques exhibit a statistically significant difference (p=0.0344, p<0.05). In terms of accuracy, the findings indicate that Novel Random Forest exceeds Logistic Regression.