- National Family Health Survey, 4 (2015-2016) - India
- Data for this project has been taken from https://data.gov.in/
- Link to the dataset: https://data.gov.in/catalog/key-indicators-national-family-health-survey-nfhs
This project explores two problems: discovering a set of attributes that predicts rate of Stunting, and analysing the influence of Contraception Usage/Awareness on attributes related to health of women and children. The NFHS-4 data has been mined using K-Means Clustering and Association Rule Mining algorithms. The results obtained related to stunting stress the importance of sanitation and health of women (during pregnancy, and overall well-being and education).
Contraceptive usage/family planning has been shown to have remarkable consequences on the behaviour of households with respect to children and mothers, children are more likely to be immunised, pregnant women are more likely to get proper health care checkups, and women are more likely to be punctilious about their health (for example, getting checkups for Breast and Cervix Cancer).
Please see the report for detailed findings.
Link to the demo video : https://www.youtube.com/watch?v=L2tnYNAUZ1c&feature=youtu.be
This project has been completed as part of the research project for CS F415 : Data Mining at BITS Pilani, Hyderabad Campus under the guidance of Dr. Manik Gupta