An essential measure of public health significance linked with infant mortality is low birth weight, a condition defined by the World Health Organization (WHO) as infants weighing less than 2500 g at birth. Some medical centers identify causes of low birth weight, but other health and demographic factors are also involved and may be directly or indirectly linked to this condition. This study aims to use prospective machine learning techniques to develop predictive models to estimate preterm birth weight using health and demographic data. Naive Bayes, Linear Regression, and Extreme Gradient Boosting are the machine learning algorithms that have been used in research for predictive analysis. The results of this study can be used by medical professionals and researchers who evaluate low birth weight babies and help the public avoid similar situations where children are born with low birth weight.