Research On The Prediction Model Of House Price Based On Machine Learning

With the rapid development of artificial intelligence technology,machine learning is widely used in all walks of life,and has achieved good results in multiple fields.House prices are a hot issue with complex influencing factors,which is difficult to make a comprehensive and accurate prediction.Therefore,this paper attempts to apply the related machine learning algorithms to house price prediction.Under a relatively stable condition,the analysis and prediction model of house prices with complex characteristics is established.This paper analyzes the current domestic and foreign house price prediction technology.It combines Internet data and machine learning technology to analyze and model house prices,respectively.what's more,it can achieve the housing price prediction.The main work of this paper is as follows:(1)Get data of house price.The research object of this paper is the housing transaction information of Chengdu,Python is used to get this data in the housing transaction website.(2)Preprocess the house price data.Statistical analysis,missing value processing,abnormal value processing,data normalization and feature selection are carried out for the original data.(3)Building the house price prediction model.The long-short term memory(LSTM)model is selected as the prediction model in this paper,and the hyper-parameters of LSTM are optimized.Because there are many hyper-parameters in the LSTM model,the adaptive artificial fish swarm algorithm is used to optimize some of them.Random forest,support vector machine,BP neural network and LSTM model are constructed as comparative experiments to predict house prices.The experimental results show that the house price prediction model based on the adaptive artificial fish swarm optimization proposed in this paper has a certain improvement in the average relative error and other evaluation indicators,the model has strong prediction ability,and can achieve the prediction of one house one price.(4)The system consists of five modules: registration and login,data display,data preprocessing,house price prediction and data import.It forms an information processing system integrating data collection,data storage,house price prediction and data display.In this paper,machine learning technology is used to explore the method of house price prediction.The model can accurately predict the house price under certain conditions.According to the model,the house price prediction system can estimate the house price and provide scientific reference for the house seller and the house buyer.