Analysis and Prediction PCOS Using Classification Algorithms

Five to Ten percent of women of reproductive age suffer with pcos, the most prevalentendocrine gland illness. Predicting the occurrence of polycystic ovarian syndrome (PCOS) is thereforea helpful method to raise women's awareness of reproductive health.In this study, machine learning and neural network algorithms are used to predict the PCOS. The siximportant classification models are applied to forecast PCOS model in women. Models werecompared and evaluated using statistical measures such as Accuracy, Balanced Accuracy, AUC-ROCCurve, and F۱ Score. Based on these evaluations, the best model was selected. AdaBoostClassifier isthought to be the best model for forecasting PCOS. The findings show that the F۱ Score, BalancedAccuracy, and AUC-ROC Curve values are ۰.۷۴, ۰.۷۴, and ۰.۶۵, respectively, and that theAdaBoostClassifier's accuracy is close to ۸۰%.

کلیدواژه ها:

polycystic ovarian syndrome ، machine learning ، Accuracy ، statistical ، AdaBoostClassifier