Application Of Machine Learning Algorithms In Total Housing And Classification Statistics

In the current era of network interconnection,with the maturity of software and hardware conditions,data collection and information storage capabilities in various fields have been greatly enhanced,but the requirements for information sorting and mining have further increased,and various industries are eager to have more variety And more efficient data mining methods to process and analyze industry data to assist the industry in making more efficient and accurate decision-making outputs.In order to meet the processing needs of big data,the field of data mining algorithms is booming,and related innovations and improvements have been born,and the compatibility of algorithm applications has gradually been demonstrated,and the application field has continued to expand.This paper mainly introduces the four algorithms of decision tree algorithm,logistic regression algorithm,combination algorithm random forest,and support vector machine algoritihm(SVM),and applies them to multi-class classification of houses.Optimize the model parameters with R software,build a suitable model,and complete predictions for missing data.From the model classification accuracy rate and the degree of difference between the predicted value and the reference value,the advantages and disadvantages of the algorithm are cross-evaluated.Finally,it is found that in the current application scenario,the decision tree algorithm has the best classification effect.The result of the reference value is also the closest.The standard deviation index is only 10.24,and the random forest algorithm has the best classification stability in the ten-fold cross-validation.The support vector machine has poor prediction stability,but it will achieve more in certain scenarios.Accurate estimation results.