Fusing Neural Networks, Genetic Algorithms and Fuzzy Logic for Analysis of Real Estate Price

It is generally acknowledged that the price of real estate are highly complicated and are interrelated with a multitude of factors. It will be advantageous if the parties to a dispute have some insights to some degree. This paper introduces a hybrid genetic algorithm (HGA) approach to instance selection in artificial neural networks (ANNs) for housing price determinants. ANN has preeminent learning ability, but BP training algorithm is based on the error gradient descent mechanism that the weight inevitably fall into the local minimum points. In this paper, an improved genetic algorithm was used to optimize the weights of neural network A case study was carried out on housing price determinants of a sample project using this model. The results concerning the efficiency of the proposed framework in terms of accuracy and computational time are also presented. It shows that more accurate price prediction of real estate can be acquired with the GA-ANN model.