Designing a Credit Scoring Model for Customers Active in the Tourism Industry with a Hybrid Approach Based on Entropy

Document Type : Original Article

Authors

1 Department of Management- Kharazmi University- Tehran- Iran

2 Department of Industrial Engineering- South Tehran Branch- Islamic Azad University-Tehran-Iran

Abstract
In recent years, credit scoring has been one of the main methods of financial institutions to assess credit risk. The main problem that limits the effectiveness of credit scoring methods is the unbalanced distribution of data, which means that in the data set, the number of samples of good customers is far more than the number of samples of bad customers. This study deals with the credit scoring of customers active in the tourism industry using an entropy-based algorithm with the aim of overcoming the problem of data imbalance, which evaluates the data in terms of the entropy of customer validation indicators and defines a criterion that can measure how good or bad a customer is by considering only the good cases of the data set and the sample of the applicants for facilities. In this research study, 204 active customers of the tourism industry of Bank Melli Iran were selected as the data set. The results showed that the entropy model has a good prediction power and is an effective model for validating customers.

Keywords