Research On The Rapid Prediction Of The Cost Of Long Rental Apartment Decoration Project

With the rise of long rental apartments,more and more enterprises join in this field.At present,most enterprisers of the long rental apartments still rely on capital investment to maintain operation.However,capital will not wait for a long time,so it is a tough challenge for those corporations to establish a feasible profit model and make profit quickly.In the long rental apartment industry,most profits come from rent and management fees,so the cost control is crucial for the enterprieses.Decoration project cost is a large part of the whole project cost,but the traditional decoration engineering estimation process is complex and cumbersome,that cannot meet the needs of the development of the long rental apartment enterprises.Based on the literatures on the cost accounting in decoration engineering and machine learning theories,the paper selected the suitable machine learning model to achieve a rapid prediction of the cost of decoration engineering.This paper first made a clear definition for long rent apartment decoration engineering cost,and researching object;then through the theoretical analysis of the system and practical experience for the combination of index system to estimate the cost of decoration engineering construction;and the selection of multiple linear regression,BP neural network and random forest of three kinds of machine learning methods,analyzes the basic system the principle of predicting decoration engineering cost by using these three methods,construct the prediction model of the new decoration engineering cost;then,the decoration project in the past,for the decoration engineering cost forecast data set;finally the new model through training and prediction data sets.The results show that the random forest model has better performance and higher prediction accuracy,and achieves a better prediction effect.In this paper,R language is used as the tool of data analysis and statistical modeling.As a new computer language,it has powerful function of graphic processing and data analysis.