| With the rapid development of the mobile Internet and the sharing economy,shared bicycles have sprung up all over the country,leading to the new modes of travel,alleviating the pressure of urban road traffic,and also becoming a research hotspot in intelligent transportation.There are many problems in the development of shared bicycle,such as the difficulty of using bicycles and parking during peak hours,the inadequate scheduling and management.However,due to the limited public data collection of shared bicycles and the relatively small theoretical research of the system,it is of practical significance and theoretical research value to mining the travel data of the shared bicycle.In this paper,the shared bicycle travel data in Beijing is taken as the research object,and the spatial and temporal characteristics of users’ travel are analyzed.Then,based on the clustering by fast search and find of density peaks algorithm,the travel hotspot areas are mined,and the LDA model is used to identify the hotspot areas.Finally,a prototype system is designed and implemented.The above research work provides reference and theoretical support for improving user experience and satisfaction,improving and guiding the management and planning services of operating companies,and promoting the construction of public service facilities such as parking areas.Specifically,this article mainly includes the following four aspects.Firstly,the data is cleaned on the shared bicycle travel data and the position data information is decoded into latitude and longitude coordinates by using the Geohash algorithm.It is necessary to obtain the travel characteristics of the user’s workday and non-workday from the time and space dimensions.The travel characteristics of the shared bicycles are interpreted in a visual way to provide a basis for the research of hotspots.Secondly,according to the space-time analysis of user travel,the clustering by fast search and find of density peaks algorithm is used to mine the starting point and the end point of the hotspots area in different time periods.Visualize the hotspot area and summarize the mining results of the travel hotspots on the working day,and provide a basis for identifying the functions of the hotspots.Thirdly,searching for POI data according to the shared bicycle data,and identifying the function of the starting point and the ending travel hotspot area of different time segments based on the LDA model.Summarize and analyze the dynamic travel characteristics of the starting and ending travel hotspots within the user’s working day.Fourth,the PyQt5 module in Python is used to design and implement a prototype system for mining and analyzing hotspots based on shared bicycle data.The system intuitively presents three functional modules:spatio-temporal analysis module,hotspot area mining module and area function recognition module. |
