Application of Radial Basis Function Neural Network Based on Ant Colony Algorithm in Credit Evaluation of Real Estate Enterprises

The credit evaluation index system is the decisive basis for investment. The existing credit evaluation systems of real estate enterprises mostly adopt the Back Propagation (BP) neural network which increasingly shows its limitations, such as slow convergent speed and easy convergence to the local minimum points. In order to evaluate the credit of real estate enterprises more reasonably and comprehensively, this paper establishes a systematic credit evaluation index system ,in which indexes, such as comprehensive qualities of leaders, loans status, third-party guarantee, have received due attention. And then, this paper proposes a new crediting evaluation model that combined ant colony algorithm (ACA) with radial basis function (RBF) neural network for quantitative evaluation, and take credit status of 30 listed real estate enterprises as samples to train and test the model. It shows not only the extensive mapping ability, but also the excellent performance of high efficiency, rapid convergence and distributed computation of ant colony algorithm. From the experimental results, it is effective and suitable to apply this method to credit comprehensive evaluation index system of real estate enterprises.