Neural network based Support Vector Machine in financial default forecast

An Artificial Intelligence based classification system will be introduced that can be helpful in separating financial ratios into two classes. The main goal was to develop a Support Vector Machine based implementation that can draw reasonably accurate conclusions even from an extensive data set. In addition, the scalability and configurability of the algorithm were important aspects too. In this article the structure of the Support Vector Machine implementation will be presented, covering the unique characteristics of the development, the problems which occurred during the construction of the model, and the solutions for these problems. The operation of the system will be presented through various tests, and it will be shown how the different parameters can describe the behaviour of the algorithm.