Ovarian Cancer Classification Using Hybrid SyntheticMinority Over-Sampling Technique and Neural Network

Every woman is at risk of ovarian cancer; about 90 percent of women whodevelop ovarian cancer are above 40 years of age, with the high number of ovariancancers occurring at the age of 60 years and above. Early and correct diagnosis ofovarian cancer can allow proper treatment and as a result reduce the mortality rate.In this paper, we proposed a hybrid of Synthetic Minority Over-Sampling Technique(SMOTE) and Artificial Neural Network (ANN) to diagnose ovarian cancer frompublic available ovarian dataset. The dataset were firstly preprocessed usingSMOTE before employing Neural Network for classification. This study shows thatperformance of Neural networks in the cancer classification is improved byemploying SMOTE preprocessing algorithm to reduce the effect of data imbalancein the dataset. To justify the performance of the proposed approach, we comparedour results with the standard neural network algorithms. The performancemeasurement evaluated was based on the accuracy, F-measure, Recall, ROC AreaMargin Curve and Precision. The results showed that SMOTE + MLP (with above96% accuracy) performed better than SMOTE + RBF and standard RBF and MLP

Ovarian Cancer Classification Using Hybrid SyntheticMinority Over-Sampling Technique and Neural Network Keywords:

Artificial Neural Network , RBF , SMOTE , MLP , Data Imbalance , Ovarian Cancer