SITE SELECTION METHOD AND DEVICE

摘要

A site selection method and device are disclosed, and relate to the field of big data analytics technologies. The method includes: obtaining spatial-temporal feature data of a plurality of users, where the spatial-temporal feature data is used to indicate actual geographical locations of the users at each moment; performing rasterizing processing on spatial-temporal feature data of each user, to convert the spatial-temporal feature data into mapped spatial-temporal feature data; determining a frequent trajectory of each user based on the mapped spatial-temporal feature data of each user; and determining, based on frequent trajectories of the plurality of users, an alternative address for a subject whose site is to be selected. Embodiments of this application provide a solution for selecting a site by extracting frequent trajectories with reference to spatial-temporal feature data of users. The frequent trajectories of the users indicate geographical locations at which the users often appear. Using the frequent trajectories for site selection can reflect a flow of people and identities of surrounding users at each geographical location, and is in accordance with an actual situation that the users are accustomed to performing various daily behaviors and activities at the locations at which the users often appear. This helps improve site selection accuracy.