The Analysis of Factors Affecting the Sustainable Livelihoods of Poverty Households using Data Mining Technique

Abstract

The purposes of this research were to classify the poverty households and to analyze the sustainable livelihood factors affecting poverty. The target group was 1,598 poverty households in Kut Bak district, Sakon Nakhon province based on Thai People Map and Analytics Platform (TPMAP). A Sustainable Livelihoods Framework was used as a conceptual framework to collect data through questionnaires, and the framework consists of five types of capital: human capital, social capital, natural capital, physical capital, and financial capital. This research applied the K-Prototype clustering algorithm to classify the groups of poverty households as well as using one-way analysis of variance (ANOVA) and Chi-Square test for statistical analysis with the significance level at 0.05.The results revealed that poverty households were divided into four categories, which were extreme poverty, high poverty, moderate poverty, and low poverty. The factors that affected the sustainable livelihoods of the poverty households were income, expenses, types of agriculture, source of the loan, water use in agriculture, the number of school-age children, the number of career skills, education levels, the number of working-age people and occupation.