Abstract:
Poverty is a condition of a person's inability to meet the standard of basic necessities of
life. Poverty is one of the indicators of people's welfare that determines the condition of
welfare and the success of development in a country. The study aims to classify poor
households in Pekanbaru City in 2020 with a total of 797 households. The classification
algorithms used are random forest and decision tree. This study shows that the random
forest algorithm produced accuracy, sensitivity, specificity, precision, and f1-score
values of 95.5%, 97.1%, 42.8%, 98.3%, and 97.7%, respectively. The decision tree
algorithm produces an accuracy value of 92.3%, 92.9%, 92.9%, 71.4%, 99.1%, and
95.9%, respectively. This proves that the random forest algorithm is the best algorithm
in classifying poor households in Pekanbaru City in 2020.