Analysis and Prediction of Polycystic Ovarian Syndrome Using Ml Classifiers

Polycystic Ovarian Syndrome (PCOS), an endocrine disorder which is common in females of reproductive age. A large number of females of reproductive age are affected by PCOS due to adverse change in their lifestyle. As of the current trend between 2.2 % to 26.7% women of childbearing age have PCOS. Visiting a gynecologist might be a taboo for many women, let alone the cost of running multiple tests. So, using medical expertise and feature selection we have filtered and obtained basic parameters that can predict if a woman has PCOS or not with minimal number of medical tests thus reducing cost and time. In this study we have used feature selection and classification algorithms to detect if a woman has PCOS or not. The algorithms used are Logistic Regression and Random Forest of which we have concluded that Random Forest provides a higher accuracy.