EARLY PREDICTION OF CERVICAL CANCER USING MACHINE LEARNING TECHNIQUES

According to recent studies and statistics, Cervical Cancer (CC) is one of the most common causes of death worldwide and mainly in the developing countries. CC has a mortality rate of around 60%, in poor developing countries and the percentages could go even higher, due to poor screening processes, lack of sensitization and several other reasons. Therefore, this paper aims to utilize the high capabilities of machine-learning techniques in the early prediction of CC. In specific, three well-known feature selection and ranking methods have been used to identify the most significant features that help in the diagnosis process. Also, eighteen different classifiers that belong to six learning strategies have been trained and extensively evaluated against primary data consisting of five hundred images. Moreover, an investigation regarding the problem of imbalance class distribution which is common in medical datasets is conducted. The results revealed that LWNB and RandomForest classifiers showed the best performance in general and considering four different evaluation metrics. Also, LWNB and logistic classifiers were the best choices to handle the problem of imbalance class distribution which is common in medical diagnosis tasks. The final conclusion which could be made is that using an ensemble model which consists of several classifiers such as LWNB, RandomForest and logistic classifiers is the best solution to handle this type of problems.