<![CDATA[Poverty is a multidimensional issue that remains a central focus of development in Bengkulu Province. Decision-tree–based approaches, particularly the Classification and Regression Tree (CART), can be used to classify households based on demographic and housing characteristics. This study utilizes data from the 2024 National Socio-Economic Survey (SUSENAS), with household poverty status as the response variable and six predictor variables including area type, gender of household head, age of household head, number of household members, floor area of the house, and housing ownership status. The analysis consists of data preprocessing, descriptive statistics, data splitting, CART model construction, identification of influential variables, and model evaluation using a confusion matrix. The results show that the number of household members, floor area of the house, and age of the household head are the most influential variables in distinguishing poor and non-poor households. Model evaluation produced an accuracy of 0.73, sensitivity of 0.58, and specificity of 0.75. The accuracy and specificity values indicate adequate classification performance, while the low sensitivity suggests that the model is still less optimal in detecting poor households, partly due to class imbalance in the dataset. These findings indicate that the CART method can be applied to poverty analysis in Bengkulu Province, although further model improvement is needed to enhance its capability in identifying poor households.]]>
Copyrights © 2025
