English  Français

A Neural Network Model to Study Non-Communicable Diseases in India

Aakanksha Agarwala, Gauhati University
Manab Deka, Arya Vidyapeeth College

Non-communicable diseases (NCDs) are the leading causes of death worldwide, and carry a huge cost that extends beyond health to trap people in poverty, deny them a life of dignity, undermine workforce productivity, and threaten economic prosperity. The present study seeks to examine the effect of socio-demographic characteristics, lifestyle behavior among people of India. We used data from Indian Human Development Survey-II (IHDS II) for the study. Neural network technique was applied to attain the study objective. Some socio-demographic factors such as age, education, caste, religion, income, smoking habit, chewing of tobacco, have been found to have significant impact on NCD. Moreover, the independent variable importance analysis of neural network technique has identified age as the most important covariate (normalized importance 100%), followed by caste, religion, income etc. The present study is an attempt to fill the void that exits when it comes to studying about the prevalence of NCDs with help of sophisticated technique like neural network.

Keywords: Health and morbidity, Spatial analysis/regression

See extended abstract.

  Presented in Session P2.