Online car-hailing passenger-carrying characteristic analysis method based on K-means

The invention discloses an online car-hailing passenger-carrying characteristic analysis method based on K-means, and the method comprises the following steps: S1, capturing an urban road image of a certain region from an electronic map; S2, performing graph segmentation on the intercepted picture based on a semi-supervised K-Means clustering algorithm to obtain a target region; S3, obtaining online car-hailing order data from the database; S4, cleaning invalid data in S3, and retaining valid data; S5, matching the effective data in the S4 with a target area, clustering the online hailed cars by adopting a K-means-based clustering algorithm, and adjusting clustering parameters according to a clustering result evaluation index to obtain an optimal clustering result; and S6, superposing the spatial distribution of the clustering result on the target area to obtain a spatial distribution diagram of the online car-hailing travel trajectory. The method has the advantages that a more accurate passenger-carrying cha