1. Classification and detection of cervical cancer for enhancement diagnosis rate using XGBoost algorithm in comparison with artificial neural network.
- Author
-
Dinesh, M. G., Dinakar, R. S., Sathish, K. S., and Venu, D.
- Subjects
- *
ARTIFICIAL neural networks , *HUMAN papillomavirus , *CERVICAL cancer diagnosis , *EARLY detection of cancer , *CERVICAL cancer - Abstract
The objective of this paper is to assess whether the utilization of the XGBoost technique, in contrast to the Artificial Neural Network (ANN) algorithm, enhances the precision of cervical cancer detection. Two distinct groups, denoted as Groups 1 and 2, were established, each comprising fifteen samples. The algorithms ANN and XGBoost were associated with Groups 1 and 2, respectively. The dataset was provided by High-risk human papillomavirus (HPV.CS), and the values for Alpha and G-power were determined as 0.94 and 80%, respectively. Employing an independent sample T-test with a significance level of p=0.007 (p<0.05), it was observed that the accuracy of the XGBoost method reached 87.333%, surpassing the accuracy of the artificial neural network approach, which was measured at 80.200%. The accuracy of the XGboost method is significantly greater at 87.333% when related to the ANN technique. It is concluded that the XGBoost algorithm outperforms the ANN by a wide margin in terms of accuracy for finding in the determination of cervical cancer. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF