This paper studied the location selection problem of pharmaceutical retail enterprises and proposed a location classification decision model based on machine learning, which provided reference and guidance for the location decision of retail enterprises and took a significant part in both theoretical and practical. We constructed a model of location selection of pharmaceutical retail based on machine learning by using the approval forms and operation data of the sample company. By processing and analyzing the historical data, constructing feature vectors and supervision information for store location selection, then the data set was divided into training data set and test data set. We trained the model and solved the optimal model parameters, validated the model on the test data set, and constructed an evaluation index to evaluate the objectivity and effectiveness of the model. The experimental results show that the decision-making model based on machine learning was effective and could be used to guide the decision-making of the location of retail stores in practice.