Credit risk analysis : a neural network approach using self organizing maps | |
| Author | Vasudeva, Vivek |
| Call Number | AIT Thesis no. CS-95-04 |
| Subject(s) | Risk Neural networks (Computer science) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering |
| Publisher | Asian Institute of Technology |
| Abstract | The subject of Neural Networks has a broad spectrum and covers a large variety of problems. One such application where Neural Networks are used extensively, as of today, is the Financial area. This study is devoted to the application of Neural Networks in the Credit Department of the banking sector. With the advent of fast computing machines, rote clerical tasks in banks began to be automated. This was followed by the automation of the accounting procedures, and transaction handling could be done in virtually no time. The need to handle repetitive analytical management decisions by the computers was soon felt. These decisions were often a result of scanning voluminous data, large amount of paper work and had to done by a skilled and experienced individual whose service fee is extremely expensive. The Loan Examiner in the credit department has to make such a decision. Automation of the credit analysis procedure, or induction of an intelligent computer, serving as an assistant to the loan examiner, who can help in making the decision of the approval or disapproval of credit, remains a dream of every lending institution. This study is done with the aim of fulfilling this dream. This study gives the design of ACES (Automated Credit Evaluation System) which can help in credit risk analysis. The core of ACES has been also implemented. ACES makes use of the Self Organized paradigm supplemented with others to achieve the goal of risk analysis. The three domains of credit analysis applications covered by us in this study are 'Credit Card Issue', 'Corporate Bank Loans' and 'Home Loans'. This system can serve as a Loan assistant, and cannot handle risk analysis single-handedly. |
| Year | 1995 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Sadananda, Ramakoti; |
| Examination Committee(s) | Huynh, Ngoc Phien;Gupta, Jyoti P.; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 1995 |