PCOS Perception analysis prediction using Machine learning algorithms

PCOS-Polycystic ovary syndrome is a prevalent disorder among women in their reproductive years, characterised by irregular periods, pelvic pain, and other symptoms like excessive hair growth on the face, stomach, thighs, and chest, excessive weight gain, and thickened skin patches. According to a large-scale national survey on PCOS, 65 percent of Indian women were unaware of PCOS signs and symptoms. This paper focuses on disseminating knowledge by developing a machine learning-based model for diagnosing whether individuals exhibit PCOS symptoms or not . As PCOS grows, a woman's ovaries and adrenal glands produce an imbalanced amount of hormones. Based on the visible changes in their bodies caused by a hormonal imbalance, this model predicts whether they should take a PCOS test or not. In this paper, we use machine learning algorithms such as logistic regression, random forest, decision tree, K nearest neighbour,naïve bayes and support vector machine (SVM) to diagnose PCOS based on network-obtained data. With the help of performance metrics like accuracy,F-statistics, recall, and precision, the performance of the algorithms was validated after an analysis of the results. The results of logistic regression with an accuracy of 90% have proven to be the most accurate prediction model out of all the possible approaches.