Comparative Analysis of Machine Learning Algorithms for Prediction of PCOS
Preeti Chauhan, Pooja Patil, Neha Rane, Pooja Raundale, Harshil Kanakia
Abstract
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder affecting many women in their child-bearing age groups. It leads to sterility and various kinds of disorders in women. The main signs of this disease are uneven mensuration cycle, obesity, oily skin, pimples, anxiety disorders. An estimated one in five women suffers from PCOS. It is seen that most women overlook the common indication of PCOS and visit the doctor only when they face difficulty conceiving. If not diagnosed in time, the condition can cause serious health issues. To overcome this problem this paper proposes to develop an application for the early prediction of PCOS using Machine Learning techniques. The required dataset is created by carrying a survey and is cleaned using Python and Google Colab. Gini importance is used to compute feature importance. Classification of PCOS is done using various machine learning techniques such as K-Nearest Neighbor(KNN), Naive Bayes, Decision Tree Classifier, Support Vector Machine(SVM), Logistic Regression(LR). Based on the accuracy and confusion matrix, the Decision Tree Classifier was found to be the most accurate model for PCOS prediction. A mobile application was developed which helps the user to predict PCOS at an early stage.