Analysis on the Growth of Shared Bike Users Based on Random Forest Model

Abstract

By analyzing the data set of hourly rental of shared bikes in Washington, D. C., this paper explores how to achieve the growth of shared bike users based on the methods of data mining and visual exploration. In this paper, machine learning models such as ridge regression, lasso regression, support vector machine regression and random forest regression are mainly selected to predict the needs of shared bike users, and then the random forest regression is verified as the optimal model. The result of this article explores the reasonable scheduling of auxiliary resources in the shared bike industry, improves the utilization rate of bicycle resources.