Abstract
Aim: The study’s aim is to analyze and compare the accuracy, sensitivity, and precision of diabetic prediction among pregnant women using the innovative Principal Component Analysis algorithm and Support Vector Analysis. Materials and Methods: This study involves two groups: Principal Component Analysis (N=20) algorithm and Support Vector Machine (N=20) with a sample size of 40 for each group. The sample size calculation uses a pre-test power of 80%, a threshold of 0.05, and a confidence interval of 95%. Results: Performance of algorithms are measured using accuracy, sensitivity, and precision. Principal Component Analysis algorithm results in mean accuracy of 79.43% significantly different with P=0.488(p>0.05), a sensitivity of 79.29% with P=0.096 (p<0.05), and a precision of 83.57%. Support Vector Machine algorithm results in mean accuracy of 77.67%, a sensitivity of 76.67%, and a precision of 83.54%. Conclusion: Principal Component Analysis algorithm performed significantly better than the Support Vector Machine algorithm for Diabetic prediction.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call