Analysis and Comparison for Prediction of Diabetic among Pregnant Women using Innovative Decision Tree algorithm over Support Vector Machine Algorithmwith Improved Accuracy

Abstract

Aim: A decision tree algorithm and Support vector machine were employed in machine learning algorithms for the prediction of diabetes among pregnant women to achieve accuracy, sensitivity, and precision.

Materials and Methods: To test the technique’s utility, researchers used open data sets such as the Pima Indian dataset from the UCI website to look at diabetes in pregnant women. This study has two groups, each with a sample size of 40: Decision tree (N=40) and Support vector machine learning (N=40). The sample size was calculated using a pre-test power of 80%, a threshold of 0.05, and a confidence interval of 95%.

Results: Algorithm performance is measured by its accuracy, sensitivity, and precision. The accuracy rate of the Decision tree is 65%, whereas the accuracy rate of the Support vector machine is 67%. The decision tree has a sensitivity rate of 54%, whereas the support vector machine sensitivity rate of 67%. The decision has a precision rate of 75%, whereas the support vector machine has a precision rate of 63%. The accuracy rate differs by a considerable amount p=0.366 with p>0.05.

Conclusion: The Support vector machine method predicts superior classifications in identifying the accuracy, sensitivity, and precision for accessing the rate for diabetes prediction among pregnant women when compared to the Innovative Decision Tree algorithm.