Prediction of Child Disease Based on Statistical Analysis of Survey Data Using Machine

Abstract

This research aims to ascertain the prevalence and the factors affecting preventive infectious diseases among small children in Bangladesh. The students assumed that morbidity occurred in infants under 5 years of age was affected by various background features of children and their parents. In Bangladesh, the death rate of infants is much higher than natural deaths. Children are dying from several diseases every day. Malaria, diarrhea and Chickenpox are most common amongst them. There were 285731 confirmed malaria cases between January 1, 2008, and December 31, 2012. Though it will be decreased because the Bangladesh Government has taken many steps against it. But still, it's a big issue. Bangladesh had the greatest percentage of deaths due to acute bloody diarrhea in children from 1 to 4 years old: 27.8 percent (5/18). Chickenpox is also a massive problem behind child deaths. On the other hand, younger children were more likely than their older counterparts to have several health problems. Children from lower-income or middle-income families were at a higher risk of disease than those from higher-income families. The prevalence of childhood moral diseases was significantly influenced by changes in drinking water supply and care methods. The 85.8 percent more likely co-morbidity in children in homes with contaminated untreated sewage is compared to parents in families with piped water. Four standard ML algorithms were used. They are- Naive Bayes Classifier, SVM, Neural Network and Random Forest Support Machines. The highest accuracy of 96.17% has been forecast by random forest.