Informatics Publishing Limited

A. Martin ¹, V. Aswathy ², S. Balaji ², V. Prasanna Venkatesan ¹

Affiliations
1 Department of Banking Technology, Pondicherry University, Puducherry, IN
2 Sri Manakula Vinayakar Engineering College, Puducherry, IN

Abstract

Many Qualitative Bankruptcy Prediction models are available. These models use non-financial information as Qualitative factors and from which Bankruptcy is predicted. In the prior researches Genetic Algorithm was applied to generate Qualitative Bankruptcy Prediction Rules. However this Model uses only very less number of Qualitative factors and the generated rules has Repetitions and overlapping. To improve the Prediction accuracy we have proposed a model which applies more number of Qualitative factors and Prediction Rules are generated using Ant Colony Optimization Algorithm (ACO). The concept pheromone depositing and updating in Ant Colony Algorithm reduce the false negative rules in the bankruptcy prediction. The heuristic and probabilistic features of Ant Colony Algorithm increase the prediction accuracy of Bankruptcy.

Keywords

Ant Colony Algorithm, Genetic Algorithm, Qualitative Bankruptcy Prediction, Repetitions and Overlapping, Qualitative Factors, Pheromone Deposit and Update, Heuristic and Probabilistic Features, Prediction Accuracy.

Full Text

     

A. Martin ¹, J. Madhusudhnan ¹, T. Miranda Lakshmi ², V. Prasanna Venkatesan ¹

Affiliations
1 Department of Banking Technology, Pondicherry University, Puducherry, IN
2 Department of Computer Science, St. Joseph's College, Cuddalore, IN

Abstract

In the globalized stiff business environment for the survival of any organization an effective technology is required to take right decisions at right time by right people. One such a prime technology is business intelligence. Bankruptcy prediction is one of the business intelligence techniques. Among so many challenges bankruptcy is very important for a financial institution or any business. Prediction of bankruptcy is crucial for the smooth running of business. Many bankruptcy models are available. Each bankruptcy model is described by quantity equation, which is based on the non linear relationship between various financial ratios used in that model. The Genetic process is applied to find the non linear relationship between financial ratios which are having more impact on bankruptcy model. In this research three bankruptcy models Altman, Edmister and Deakin model were chosen. Genetic algorithm is applied in these three bankruptcy models to find most impacted ratios. Altman model is has more impact on its financial ratios compare to other bankruptcy models. The impacted threshold value is 98% matches with the original threshold value of Altman.

Keywords

Genetic Algorithm, Bankruptcy Models, Deakin Model, Altman Model, Edmister Model, Financial Ratios, Business Intelligence.

Full Text