Abstract:

Abstract PCOS is the reproductive metabolic condition in which the ovary produces the number of follicles that is unusually high. The number of follicles, size, and location of the ovary are observed using the data set of the ovary. Because of the varied sizes of follicles and the fact that it is strongly linked to veins and tissues, radiologists have traditionally had a tough time diagnosing PCOS. To predict the PCO syndrome fertility and infertility for the data collected the from KAGGLE repository, preprocessing techniques are being used to extract useful information for analysis. Heat Map is the preprocessing technique used for identifying correlated features. Then the extracted data are considered for training and testing to classify the occurrence and nonappearance of PCO syndrome. For data training and classification, Support Vector Machine, KNN, Naive Bayes, and Hybrid Algorithm are used. The proposed approach outperforms other current methods and has been proven to be effective.

Keywords: