Background <p>Cervical cancer poses a significant public health challenge, particularly in India where it accounts for a considerable number of cases and deaths. Despite the effectiveness of early screening in preventing deaths, a large percentage of Indian women are diagnosed in advanced stages. This study uses machine learning to understand the factors influencing cervical cancer screening in India.</p> Methods <p>Utilizing data from the National Family Health Survey (NFHS 2019–21), this study analyzed the cervical cancer screening status among women of early middle-aged women (30–45) across different districts in India. A range of independent variables, including sociodemographic and behavioral factors, were examined. The study employed machine learning techniques, specifically Weighted Random Forest, to identify key factors affecting cervical cancer screening uptake.</p> Results <p>The results show that age, parity, wealth index, and education level are the most influential predictors of screening behavior, with higher feature importance values. The WRF model produced 94.4% accuracy with a ROC-AUC value of 0.707.</p> Conclusion <p>Weighted Random Forest gave a methodological benefit by effectively addressing class imbalance and improving forecast accuracy for screening participation. This model enables the reliable identification of key sociodemographic predictors, providing valuable insights for targeted public health interventions that support policies aimed at improving cervical cancer screening access, education, and affordability, particularly among underserved populations.</p>