Credit Scoring Model for Farmers using Random Forest

One of the problems faced by farmers in Indonesia is capital. Based on Indonesian Central Statistics Agency survey results, number who borrow capital from formal institutions such as banks still small. This because process applying for loans at lengthy, are considered high-risk and unbankable, rating agricultural sector unattractive to banks. study aims determine attributes design a model credit assessment. uses secondary data related bank ratings land productivity Telagasari sub-district 2018–2020 Cipayung 2020. Data were analyzed using random forests. The research includes four stages: collection, pre-processing, building, analysis evaluation. produced five important variables that relevant farmers: planting costs, sales, productivity, total production, area. built produces most optimal accuracy 83% with an AUC score 81%. performance classification, it can be concluded has been made good predicting status value included classification predicate.