Research On The Real Estate Appraisal Of Surrounding School On The BP Neural Network

With the continuous development of society,parents pay more and more attention to their children's education.In order to make their children gain early advantages in the future competition,it has become the consensus of parents to get excellent education opportunities in the primary education stage.There is a general imbalance of basic education resources in every city of our country.Excellent education resources are often concentrated in a few key primary schools.Therefore,entering the key primary school has become the consistent pursuit of parents.With the implementation of the nearby enrollment policy,the houses around the high-quality education resources are favored by parents and become the so-called"School District Housing”,together with other school selection channels are gradually tightened,"choosing schools by housing" has become the only choice for parents who attach importance to education.The demand for education drives the housing price of the school district to keep rising.The impact of basic education resources on the housing price is becoming more and more obvious.The capitalization of basic education resources in different degrees which has attracted the attention of scholars.BP neural network can simulate the learning and thinking ability of human brain.It is a powerful and widely used machine learning algorithm,which is widely used in the field of value evaluation.Its adaptability,nonlinearity and large-scale parallel processing ability enable it to reduce human and material resources,deal with nonlinear problems efficiently,reduce subjective randomness,and greatly improve the efficiency and accuracy of evaluation.This paper first explains the reasons for the emergence of school district housing and why it is necessary to study the value of school district housing,and then describes the current research situation of domestic and foreign scholars in the evaluation of school district housing price and artificial neural network,and then puts forward the research ideas and research ideas of this paper.Secondly,from the perspective of traditional real estate evaluation,it introduces the types and value points of traditional real estate,and analyzes the differences between school district real estate and traditional real estate value evaluation.Then,after analyzing the advantages and disadvantages of the traditional real estate evaluation methods,the paper puts forward to apply BP neural network to the school district housing price evaluation,and introduces the BP neural network and its implementation steps in detail to verify the feasibility and superiority of using BP neural network.In the process of evaluating the value of school district house price,the hedonic characteristic price model is introduced to modify the evaluation model.Using hedonic characteristic price model to optimize the sample index system,remove the influence of collinearity between the indexes,and improve the accuracy of the estimation model.The optimized sample is introduced into BP neural network as the input quantity,which can train itself,establish nonlinear map,and finally accurately predict the value of real estateIn this paper,taking Donghu District,Xihu District and Honggutan New District,where education resources are concentrated in Nanchang City,as the research scope,Python is used to collect the qualified data of anjuke and Nanchang real estate information network,and these data are screened and processed.Using MATLAB to build a value evaluation model,through case analysis and calculation,it is found that the model can accurately predict the value of school district housing,which proves the accuracy and effectiveness of this improved evaluation methodFinally,due to the limited level of the author,there are still some deficiencies in this paper,for example,some indicators will have little impact on house prices.In order to reduce the difficulty of evaluation,we can only delete this indicator,and some special circumstances such as information asymmetry between the two sides of the transaction lead to the transaction price deviate from the average line.We hope that we can find a way to quantify these special circumstances in the future;and most of the real estate in this paper The product price data is the listing price,and a small part is the transaction price on the real estate information network,because there is a certain illegal risk in crawling the data on the real estate information network,which also has a little impact on the accuracy of the assessment.I hope that the Nanchang municipal government can set up the real estate transaction database in the future,so that we can study the real estate market in Nanchang more accurately,and improve the development of the real estate industry in Nanchang and make more valuable suggestions.