Predicting Financial Assistance Requests using the Social Map Survey
Conference proceedings article
Authors/Editors
No matching items found.
Strategic Research Themes
Publication Details
Author list: Tulyathan A., Samasiri P.
Publisher: Elsevier
Publication year: 2021
Start page: 48
End page: 52
Number of pages: 5
ISBN: 9781665428415
ISSN: 0928-4931
eISSN: 1873-0191
Languages: English-Great Britain (EN-GB)
Abstract
A survey was conducted in a vast scale across Thailand during 2018 regarding the Social Map project, a Big Data initiation project led by the Ministry of Social Development and Human Security. The Social Map (SM) survey data aim to gather information on social problem encountered by the Thais. One of the questions include their requests for financial assistance from the authorized government bodies. This study has developed a predictive model on the financial assistance request prediction up on an individual demographics data. Three classification modelling approaches have been explored and XGBoost was found to give the most reliable predictive results. The model precision is in the range of 0.7-0.8 depending on the criteria used. Feature importance analysis can help identify the factors that affect the tendency for an individual to be in needs of the financial assistance. These factors include being a farmer, age, household income, number of children, number of household members and age of the householder. Later, we illustrate that the model may be applied to other data sets to reveal further useful insight which can be useful for policy makers initiate actions that can target those vulnerable groups more precisely. © 2021 IEEE.
Keywords
social map, vulnerable groups
Contact Information