Determinants of Childhood Vaccination Uptake: A Machine Learning Approach Using a Decision Tree Classifier

This study applies a Decision Tree Classifier to determine the most influential factors affecting childhood vaccination uptake in Tanzania. Methodologically, the study was quantitative in nature, using the power of Decision Support Classifier in decision making. As a result of applying a Decision Tree Classifier, six key predictors were identified as influential to the model: birth interval, number of antenatal visits, maternal age, child age, number of births in the last five years, and number of children under five years old. Among these, birth interval (33%) and antenatal visits (13%) had the highest impact on vaccination uptake. The findings indicate that shorter birth intervals, fewer antenatal visits, and younger maternal age reduce the likelihood of child vaccination. Additionally, children below 18 months and those from large households are more likely to miss immunisation. The study underscores the importance of family planning, antenatal education, healthcare access, and targeted immunisation campaigns to improve vaccination rates. These insights provide data-driven recommendations for policymakers and healthcare professionals to enhance childhood immunisation programs and reduce preventable child mortality.