Saturday , June 23 2018

SADM – An Automated System Based on
Data Mining for Credit Scoring

Petroleum-Gas University of Ploieşti
39 Bucureşti Blvd, Ploieşti, 100680, Romania

Abstract: The credit approval represents a complex banking process which requires a significant involvement of the people and departments responsible with the final decision. The decisions taken by the Central Biro as a result of credit analysis may be alsoaffected by the subjectivity of the working people in charge, causing the decision-making quality. In order to facilitate the credit approval process, the automation of the credit function made by an automated system may represent a solution in this case. In this paper an automated system based on data mining techniques is proposed and designed to assist the decision credit process. The strong points of this prototype consist in increasing the accuracy of credit decision, increasing the bank’s profitability, decreasing the credit risk etc.

Keywords: automated system, data mining, credit scoring.

>Full text
Irina IONIŢĂ, SADM – An Automated System Based on Data Mining for Credit Scoring, Studies in Informatics and Control, ISSN 1220-1766, vol. 22 (4), pp. 291-298, 2013.

  1. Introduction

The banking domain represents a dynamic area with processes that need a significant attention from the managers. Knowledge has become more and more appreciated by organizations because of the value offered to manage the entire activity of them. In a competitive world banks have to apply strategies to assure their continuous functionality and to increase their profit. Most of the knowledge in the banking system is currently generated by daily transactions and operations. The repository of data contains an enormous volume of records than hide valuable information for banks. An important task in this case is to discover that kind of information with significant implication for bank’s management. Data mining as an artificial intelligent technique appears to solve this problem. Mining data is similar to the process of mining gold and supposes a difficult work, strong algorithms and methods to satisfy the quality level of the knowledge discovered.

The data generated by the bank’s information systems, manual or automated such as ATM’s and credit card processing, were designed to support or track daily transactions, simultaneously satisfying internal and external audit requirements, and meeting government or central bank regulations. Before data mining can proceed to find nontrivial information from large databases a data warehouse will have to be created first.

Data warehousing is known as the process of extracting, cleaning, transforming, and standardizing incompatible data from the bank’s current transaction systems so that these data can be mined and analyzed for useful patterns, relationships, and associations. Data mining is applied with success to prevent attrition, to cross sell and to do target marketing, to detect and deter fraud, to prevent defaults, bad loans, to increase loyalty and customer retention etc.

A challenge for the authoress of this article was to consider the bank as an automated control system, combining the Automatics theories with banking regulations and artificial intelligence techniques in order to implement a prototype of an automated system based on data mining for credit scoring.


  1. GORUNESCU, F., Data Mining. Concepte, Modele şi Tehnici, Editura Albastră, Cluj-Napoca, 2006.
  2. HAN, J., M. KAMBER, Data Mining Concepts and Techniques, San Francisco, Academic Press, 2001.
  3. IONITA, I., IONITA, L., A Decision Support based on Data Mining in e-Banking, Roedunet International Conference (RoEduNet) 2011 10th, 2011, pp. 1-5.
  4. ILAS, C., Teoria sistemelor de reglare automată, Bucureşti, Matrix Rom, 2006.
  5. LIU, Y., A Framework of Data Mining Application Process for Credit Scoring,
  6. KISS, F., Credit Scoring Processes from a Knowledge Management Perspective, Periodica Polytehnica Services, Social, and Management Sciences, vol. 11, no. 1, 2003, pp. 95-110.
  7. MEGGS, G., M. FAGAN, Knowledge Discovery – Practical Methodology and Case Studies, Tutorial Notes in PADD98 – 2nd International Conference on Practical Applications of Knowledge Discovery and Data Mining, London, 1998.
  8. MIHALACHE, S., Elemente de ingineria reglării automate, Bucureşti, Matrix Rom, 2008.
  9. PARASCHIV, N., G. RĂDULESCU, Introducere în ştiinţa sistemelor şi a calculatoarelor, Editura Matrix Rom, Bucureşti, 2007.
  10. SCHOMAKER, C. A. M., Credit Scoring. An Overview of Traditional Methods and Recent Improvements, BMI Paper, December, 2006.
  11. SIDDIQI, N., Credit Risk Scorecards. Developing and Implementing Intelligent Credit Scoring, John Wiley &Sons Inc., New Jersey, 2006.
  12. WANG, X. Z., Data Mining and Knowledge Discovery for Process Monitoring and Control, Advances in Industrial Control, Springer, 1999.
  13. WEISS, S. M., N. INDURKHYA, Predictive Data Mining: a Practical guide, Morgan Kaufmann Publishers, San Francisco, California, 1998.