A Study on Determining Household Poverty Status: SVM Based Classification Model

Abstract

Poverty is the normal challenge faced by the worldwide community. The human society has never ceased to fight against poverty. This research study focuses on determining significant attributes that can be utilized to distinguish poor and non-poor household units. At least one selected community in the Philippines is utilized to validate and test the model for Classification using Support Vector Machine (SVM) algorithm. To check the accuracy and evaluate the model 80% of the total poor and non-poor households are used as a training set and the remaining 20% as a testing set to minimize the impact of disparities and determine whether the model's classifications are correct. Accuracy, Precision, Recall and F1-Score are likewise done to interpret and gauge the performance of the SVM algorithm for the binary classification model in which the outcome indicates 88.64% precise.