An Effective Prediction of Cervical Cancer with Hybridization of K-Means with RFT Algorithms

The cervical cancer is the second-deadliest illness for women worldwide. Cervical cancer is hidden in women until it is advanced and incurable, thus numerous trials and research projects have been conducted to find it early. Due to the lack of appropriate datasets, accurate cervical cancer prediction has proven to be the most difficult and complex task in medical data mining. Numerous studies have been conducted to provide various methods that can fix issues and increase the accuracy of cervical cancer prognosis through photographs. As a result, researchers face a difficult and time-consuming procedure that necessitates a highly effective and complex algorithm that draws from a wide range of knowledge, experience, and methodologies. At first, the Cervical Cancer dataset is completely loaded, including its primary properties and the origin of the data. After a new method termed “whitening” is applied as pre-processing to the data set, the data mining algorithm RFT is used to clean up the data. After explaining why a combination of machine learning and data mining is necessary, the paper presents the proposed technique. The problem is solved, and the algorithm is defended as the most effective one for predicting cases of cervical cancer. The most effective prediction method is a hybrid of RFT and K-Means. To aid the researcher in drawing conclusions about the optimal algorithm to utilise for reliable prediction, several algorithms are constructed and their accuracy levels are projected using sensitivity, specificity, and F-Score as prediction factors.