Attribute Set Reduction Using Filter Selection Method For Machine Learning Model To Diagnose Malnutrition In Preschool Children
DOI:
https://doi.org/10.64252/0mfkvr44Keywords:
Malnutrition Detection, Preschool Children, Machine Learning, EDA, Feature Selection, Filter Method.Abstract
A machine learning model built to detect malnutrition in a preschool child may aid parents in early detection and timely intervention for recovering child’s health, especially in situations where experts’ consultation is not available. It becomes challenging to diagnose malnutrition in a preschool child based on simple health attribute values. The article explains the steps undertaken for attribute selection playing major role in the malnutrition detection.
METHODS: Based on a sample data collected for preschool children with the help of medical experts different statistical algorithms were executed to find the contribution of each attribute in the detection process. The majorly used techniques were analysing and understanding data using different graphical representations and correlation.
Symptoms of malnutrition in a preschool child using Machine Learning algorithm is a quite challenging task which will work and in absence of expert, will help in the early detection of type of Malnutrition the child is suffering with. The process follows the steps defined as Exploratory Data Analysis (EDA), which is considered to be a crucial stage in development of any machine learning model.
