Get-off point recommendation method and device based on specific scene

Abstract

The invention discloses a get-off point recommendation method based on a specific scene, and the method comprises the steps: obtaining a get-off point recommendation request initiated by a target passenger terminal, and the get-off point recommendation request carries the position information and departure time of the target passenger terminal; identifying a vehicle use scene associated with the departure time; constructing a prediction model based on the historical car orders of the associated passenger terminals, wherein the user of the associated passenger terminal and the user of the target passenger terminal have an association relationship, and the historical order comprises passenger information, historical get-off points, historical common travel moments and travel frequencies of the historical get-off points; inputting the position information, the departure time and the vehicle use scene into a prediction model to obtain a plurality of high-probability recommended getting-off points; and pushing the plurality of recommended get-off points to the target passenger terminal. According to the invention, through the historical car use data of the user associated with the target passenger terminal, the interested places of the user are intelligently recommended in a specific scene.

Description

technical field [0001] The invention belongs to the technical field of network car-hailing alighting point recommendation, and in particular relates to a method and device for recommending an alighting point based on a specific scene. Background technique [0002] Alighting point prediction refers to predicting the destination of the user's current trip under the condition of knowing the user's location and travel time. Destination prediction is widely used in many situations. For example, in the taxi-hailing scene, when the user logs in to the taxi-hailing platform, the taxi-hailing platform will predict and recommend the user's current travel destination, which can help the user quickly take a taxi and effectively improve the user's The ride-hailing experience will bring more benefits to the platform. [0003] In the related technology, the prediction of the destination mainly uses the prediction model to predict the historical order data of the passengers. The predictio...

AI Technical Summary

Problems solved by technology

[0004] If the user travels to this destination less frequently, the model modeling cannot be satisfied, and the probability of passengers traveling to this destination cannot be predicted

On the other hand, the existing technology recommends the alighting point based on the passenger's own historical records, and the recommendation of the alighting point is less considered, and the intelligent recommendation effect cannot be achieved.

Images