<![CDATA[Cervical cancer is a leading cause of death among women, with over 300,000 deaths recorded in 2020. This study aims to improve the accuracy of cervical cancer diagnosis classification through a combination of Adaptive Synthetic Sampling (ADASYN) and Random Forest algorithm. The research data was obtained from the Cervical Cancer dataset in the UCI Machine Learning Repository with an imbalanced data distribution of 95% negative class and 5% positive class. ADASYN method was chosen for its ability to handle imbalanced data by focusing on minority data points that are difficult to classify. The Random Forest algorithm was optimized using GridSearchCV to achieve maximum performance. Results show that this combination improved accuracy from 96.5% to 96.8% and recall from 93.7% to 94.3%. Feature importance analysis identified key risk factors such as number of pregnancies, age at first sexual intercourse, and hormonal contraceptive use that significantly influence diagnosis. This research demonstrates the effectiveness of combining ADASYN and Random Forest in enhancing classification performance for early cervical cancer detection.]]>
Copyrights © 2025
