Good Food Score Reflecting Health Status of the Families: A Predicting Approach through Machine Learning

Dr. Hina Ali; Hina Shafiq; Zinaz Aisha; Maria Iftikhar; Nazish Iftikhar; Hira Ali

Authors

Dr. Hina Ali Associate Professor, Department of Economics, The Women University Multan
Hina Shafiq Assistant Professor, Department of Economics, Sardar Bahadur Khan Women's University Quetta
Zinaz Aisha Assistant Professor, Department of Economics, Sardar Bahadur Khan Women's University Quetta
Maria Iftikhar PhD Scholar, Department of Economics, the Women University Multan
Nazish Iftikhar PhD Scholar, Department of Economics, the Women University Multan
Hira Ali Assistant Professor, Department of English, the Women University Multan, Pakistan

Keywords:

Household health status, good food score, household socio-economics characteristics, logistic regression, Southern Punjab

Abstract

The current scenario investigates the different factors that affect the Household Health Status (HHS) in Southern Punjab Pakistan. Household Good Food Score (GFS) indicate a family’s financial access to a variety of foods to meet their nutritional requirements. The primary source of information was 300 instances of households. A particularly coordinated overview for get-together responses was made according to FAO rules. The Weka classification for logistic regression was used to analyze the data. The cross validation estimated coefficients of the GFS, household head income, education level, and covid-19 status all have positive effects on HHS while the attribute household head age and size has a negative effect. The findings of that study indicate the correctly classified instances that out of 300 instances, 298 instances are correctly classified. So, 99% of that is the accuracy of this particular classifier. So, the value of the kappa statistic is 0.99 and it is considered a very good value. The study also suggested key policy recommendations to improve HHS and socioeconomic factors that affect households’ health scores.