Ovarian cancer is a disorder of ovarian cell growth that is triggered by series of acquired mutations affecting a single cell or its clonal progeny. It is purposeless prey on host and virtually autonomous. It is usually diagnosed at a late stage because of poor sensitivity of screening test. There are still no effective cures for this illness. Still early detection might lower the mortality rate. Our project's major goal is to conduct predictive analytics for early detection by using machine learning models and statistical techniques on clinical data collected from specific patients. Mutual information testing is crucial in statistical analysis for identifying indicative biomarkers. A collection of machine learning models, such as the Random Forest (RF), Extreme Gradient Boosting Machine (XGBoost), Logistic Regression (LR), Gradient Boosting Machine (GBM), and Light Gradient Boosting Machine (LGBM)are utilized in the classificationof ovarian tumors as benign or malignant. By using proposed system, it can significantly identify the class of benign and malignant patients. The data collected is analyzed and pre-processed before it is used for model training and testing.