Application of Random Forest Algorithm on Credit Risk Analysis

Abstract

Credit risk is a problem of the inability of debtor to pay bills that can cause losses to both parties concerned. In order to avoid losses caused by bad loans, the loan provider have to conduct an analysis on the debtor. The loan providers are required to be objective and accurate in evaluating credit application. During the credit application, the debtor has to submit data such as personal data, monthly income, employment history, and other data. The data is then proceeded to generate credit score, based on the score the credit application will be accepted and rejected. The screening process must be done quickly and accurately. The machine learning approach is suitable for overcoming the problem of timing and accuracy of predictions and can predict factors that affect loan approval with feature importance. This research aims to identify the features that have a high feature importance with a machine learning approach using the Random Forest Algorithm. The stages of the research method used are data understanding, feature extraction, data pre-processing, exploratory data analysis, modeling, and insight. The metrics used to evaluate the model performance is recall, this model produces recall score of 0.9091. Based on research, there are some most important factors that affect loan approval, repeat_user, mobile_likelihood, monthly_salary, industry, year_of_service

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.