<![CDATA[Cervical cancer data is high-dimensional with numerous features. Reducing the number of features in the analysis can provide advantages in more effective data processing and prevent overfitting, which can lead to detection errors. This study aimed at improving the classification performance for early identification of cervical cancer through the CFS technique based on the Naive Bayes classification algorithm. The dataset used was the primary data with three classes of cancer conditions. Feature reduction was applied to decrease data dimensionality and improve processing efficiency. The selected feature subset comprised 10 attributes that showed a strong correlation with the target class. The performance evaluation yielded an accuracy of 83.50%, recall of 83.57%, and precision of 88.30%. These findings suggest that the proposed method can enhance the early detection of cervical cancer, which support early detection of cervical cancer and assist clinical decision-making]]>
Copyrights © 2026
