A functional clustering method for public bicycle rental points

Abstract

The invention discloses a function clustering method of rental points of a public bicycle system. The present invention comprises the following steps: Step 1. For the public bicycle system, first use the location information of the rental points, and use the Thiessen polygon algorithm to divide the city into regions. Each rental point covers a certain area, and each rental point is regarded as a " text" for mining analysis. Step 2. Convert the usage history of the public bicycle system into the "words" contained in the rental point. Step 3, use the LDA algorithm and the K-means algorithm to identify the rental point function of the rental point. Step 4. Analyze the spatio-temporal characteristics of each clustering result, and define the functions represented by the rental points included in each clustering. The positive effects of the invention are: provide basis for future planning of public bicycles; guide users to use public bicycles; can guide the establishment of commerce and advertisements, and have good industry application prospects.

Description

technical field [0001] The invention belongs to the field of public bicycle system planning in traffic planning, and in particular relates to a functional clustering method of public bicycle system rental points. Background technique [0002] Functional identification of urban areas refers to the use of urban operating data and dynamic features to identify and label different areas of the city according to their functions in the city. It is the main component of urban planning, which can help people understand the internal development law of the city, carry out urban planning, and guide the development of urban construction. However, it is very difficult to find this kind of research on urban functional areas: the activities of urban residents are complex, and urban functional areas are not single, so it is difficult to accurately define the specific functions of an area; traditional methods are mostly user research and field statistics. It is very time-consuming and labor-...

AI Technical Summary

Problems solved by technology

However, it is very difficult to find this kind of research on urban functional areas: the activities of urban residents are complex, and urban functional areas are not single, so it is difficult to accurately define the specific functions of an area; traditional methods are mostly user research and field statistics. It is very time-consuming and labor-intensive; and the recently adopted research object - urban data has a wide variety, large scale, multi-granularity, multi-dimensional and other characteristics, it is very difficult to find the law from it; because the urban functional area itself cannot directly Observation, so it is difficult to verify the validity and correctness of the algorithm

Images