Improved performance of machine learning algorithms for prognosis of cervical cancer

With the progression of artificial intelligence in medical services, the world has achieved many benefits. The constant improvement of existing artificial intelligence techniques becomes a boon in the medical field for assisting healthcare providers. In current years, the diagnosis of cancers using machine learning techniques for timely decisions has gained popularity. Cancer is preventable and can be cured with early and timely diagnosis. Cervical cancer is one of the foremost cancers in other female cancers which ranked at the fourth position. The objective of this study to develop a model that provides a timely and cost-effective cervical cancer risk prediction score by using supervised machine learning techniques in amalgamation with dimensionality reduction techniques. The dimensionality reduction techniques help in providing the prediction with a minimum number of features. The experimental investigation on cervical cancer risk factor reveals that Random Forest classifier using recursive feature elimination with cross-validation technique gives 93%.