High rates of maternal mortality, infant mortality, and preterm births, as well as continuing disparities in pregnancy outcomes are an important public health issue in India and worldwide. Health care industries must focus on improving the quality of treatment and continuing care of pregnant women. This study aims comparison of logistic regression with data mining techniques to identify most influenced predictor variables and to develop a decision support system to help the physicians for better decision making in low weight child birth. It is identified that the variables which were highly influenced to predict the low weight child birth are Mother’s last weight (pounds) before becoming pregnant, Mothers age, Number of physician visits during the first trimester, Number of previous premature labors. The results of this work have improved prediction accuracy in Datamining techniques when compared to logistic regression.

Key words

Low birth weight, Healthcare, Data mining, Logistic regression, Classifier accuracy.