Mathematical and intellectual methods of assessing the creditworthiness of individuals

The methodology of assessing borrowers' credit solvency is being considered and improved by building a combined system of intellectual and economic-mathematical methods. Various methodologies and tools used to assess the creditworthiness of individuals are being investigated: data mining methods, fuzzy algorithms, economic and mathematical methods. In the language of functional modeling, a description of the process of recognizing the client's creditworthiness level is presented, with the help of which the main stages of assessing the credit solvency of clients and their maintenance are determined. As part of the study of the instrumental apparatus, intellectual and economic-mathematical methods have been studied in detail, which can be synthesized into a single system for assessing the credit solvency of borrowers. Based on the fuzzy modeling tools, an algorithm is proposed that allows calculating a complex indicator of the creditworthiness of individuals. The credit solvency rating is determined on the basis of a set of criteria and features that characterize the bank's borrowers with a sufficient degree of completeness. For each selected criterion, linguistic variables “Indicator level” and classification tables are formed, correlating the values of the criteria with its “level”. The analysis of software tools used to assess credit solvency was also carried out, and the choice of an analytical platform for our own practical experiments was justified. On the basis of the Deductor platform, the scoring method, the decision tree method are implemented and a neural network is built. To implement the task of assessing the credit solvency of an individual, the concept of developing an intelligent system is proposed. The individual results of the study and the conclusions obtained can be used by specialists of credit and financial institutions to calculate the rating rating of the borrower. The systematized knowledge can be used in further research on this issue and form the basis for the implementation of the software product.