Empirical Analysis of Factors Influencing the Quality of Land Grade for Collective Management Construction Land Based on Random Forest Regression Model
Using the project of grading and benchmark land price evaluation of collective construction land at the local level in Nanchang Cit-y, we obtained survey data from 258 administrative villages in the jurisdiction of Nanchang City,and random forest regression model was used to analyze the influencing factors of land quality for rural collective commercial construction land and the degree of importance of various fac-tors, while also verifying the accuracy of the model. The results showed that the model accuracy and decision coefficient (R2) were greater than 80％, and the accuracy of the 5-fold cross validation results was also greater than 80％. The overall importance of factor layer was that macro location influence degree, prosperity level, industrial agglomeration effect, traffic conditions and regional planning factors were greater than infrastructure conditions and environmental conditions. Among them, the urban area was most affected by the industrial agglomeration effect factor, while the suburban area was most affected by the traffic planning condition factor, and both were greatly affected by the prosperity degree and traffic condition factor.The overall importance of the factors layer was that macro location influence degree had the greatest influ-ence, followed by bustling degree, traffic conditions, industrial agglomeration effect and regional planning, while the remaining factors had rel-atively less influence. Among them, the urban area was most affected by the industrial agglomeration effect, while the suburban area was the most affected by the regional planning conditions. Secondly, the prosperity degree and traffic conditions also had a greater impact on the quali-ty, while the macro location influence degree was relatively reduced.
Collective commercial construction landLand grade qualityInfluencing factorsRandom forest regression model
万方数据
维普
