Urban building planning method based on machine learning

The invention discloses an urban building planning method based on machine learning, and the method carries out the feature extraction of an existing business center case. And then PCA algorithm processing optimization is carried out on the characteristic values. A label value 1 or-1 is given according to the business state of the existing business center; and the data set is subjected to training prediction of a logistic regression model, and final model training can be completed when the confidence reaches a relatively high value. And finally, carrying out real-time feature extraction on the position and location of a current to-be-planned business center, inputting the feature into the model, and judging whether the business center can be established at the position or not, namely whether profit can be made or not after establishment. According to the method, a large amount of data can be learned and analyzed, the action relationship among various factors can be accurately predicted, and the scientificity