An Automatic Method for Identification of Cervical Cancer based on Multilayer Perceptron Neural Network

Around the world, cervical cancer is the leading cause of death among women in their reproductive years. Screening and early diagnosis are both important preventive measures that can lead to improvements in cervical cancer treatment. The machine learning methodology has the potential to process enormous volumes of information relating to cancer, and it is easily available to the medical research community. This could result in an improved patient survival rate. In the current study, the Gradient Boosting Classification was utilized to make predictions regarding the symptoms for several cervical cancer diagnostic exams. When compared to the selection tree and the Random Forest classifier, the proposed approach demonstrated much higher levels of reliability in its results. To make the model more effective in the future, we are going to have to switch to using a multilayer perceptron.