Polycystic Ovarian Syndrome Analysis using Machine Learning
Nishithha Premkumar Sangeetha, Neshanthini Vengatesan, Kavitha Prithiviraj, Maheswari Marimuthu · 2024
The most significant issue affecting the female population is PCOD (Polycystic Ovary Disorder) or PCOS (Polycystic Ovary Syndrome). It’s among the biggest issues that women nowadays are dealing with. It is a hormonal imbalance that makes the ovaries to expand, holding a little blister over the periphery. During the reproductive years, the hormonal imbalance only manifests itself. Female with polycystic ovarian syndrome (PCOS) having issues in regulation of androgen system incorporates the study of PCOD, which comprises stages like data collection, data analysis & visualization, training of the models, evaluating the models, comparison, and selecting the models, and ensemble model. The data is taken from Kaggle, processed, then visualized using visualization metrics. The model is then trained using machine learning approaches such as KNN, Naive Bayes, SVM, Decision Tree, Logistic Regression, and Random Forest. Performance measurements like confusion matrix, precision, recall, f1 score, support, and accuracy are used to assess the trained models’ correctness and efficiency. Based on these performance metrics, Decision Tree and Random Forest integrated into a single model utilizing ensemble learning had a 76% accuracy rate.