You have no items in your shopping cart.
ABSTRACT
Cervical cancer is the fourth most dangerous disease in the world, the challenge with its prediction and treatment are numerous which includes the scarcity of medical experts to attend to patients with suspected cases, the issues with clinical results variability and sundry limitations. This project work is motivated to design an improved non- invasive approach that uses machine learning and other statistical techniques for the prediction of cervical cancer. The proposed model is based on the combined use of both feature selection conducted on the dataset to distinctively select the most relevant features to aid in maximizing the learning algorithms. Also, the ensemble technique was modelled to further improve on the classification performance of our study. The work allows improvement with a higher accuracy compared with the existing works. The study also show room for future improvement in using other modelling techniques and using other advanced feature selectors.
