Data Mining Based Balancing Prediction Of Shared Bicycle

After more than two years of rapid development,the sharing of bicycles has exposed a lot of operational problems while bringing convenient services to the public: cycling stops,some areas of the time a bicycles is difficult to find,sharing bicycles "Centennial Cemetery" and so on.In view of the existing problems,control operations from the source,efficient and reasonable cycle delivery level can effectively reduce operating costs,reduce the disorderly delivery of new cars,uneven distribution of bicycles and other issues.Shared cycling the regional balance prediction problem consists of 2 parts: time balance and geographical balance.Taking the motorcycle in Taiyuan as an example,this paper mainly studies and finds the problem of motorcycle placement in Taiyuan City,and makes predictive modeling according to the anomaly detection and cluster analysis.Using Python web crawler to obtain the location information data of the motorcycle in Taiyuan City,and to predict the data visualization analysis,combined with the machine learning method to forecast the regional balance of the motorcycle in Taiyuan,for the fine operation of the enterprise,the delivery of the sharing cycle to do a good foundation.The research contents of this paper are as follows:(1)The use of Python network Crawler to obtain the motorcycle bike dataset,combined with the Python language in the strip,Split function to extract the dataset,stored in the MySQL database,combined with MySQL database characteristics to the dataset to re-weight.(2)Combining with the GAO Tak map API to visualize the data of the motorcycle bicycle,this paper analyzes the distribution heat attempt of the motorcycle area in Taiyuan City,the utilization rate of the change with time,the usage and the distance distribution of the riding vehicle.(3)Using clustering algorithm in data mining to divide the area of motorcycle cycling in Taiyuan City.Combining the distance modeling between the motorcycle area and the nearby business Circle,the unbalance of the motorcycle area is observed according to its proximity to the location of the business circle,and the equilibrium prediction of each motorcycle cycle region is obtained.Based on the analysis of the riding data of the morning time of the motorcycle in Taiyuan City,this paper finds that there is a positive proportional relationship between the imbalance of bicycles in each region and the distance of the commercial circle,and finds that the number of regional bicycles is divided into three categories with the characteristics of time: stable change,linear change,completely random change.In the bicycle delivery operation,has a high reference value.