Patent · US Active

Systems and methods for generating personalized destination recommendations

US11011057B2 · kind B2 · utility

0Cited by

1References

20Claims

0Family size

Assignee

BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. · CN

Inventors

Ran Chen · Fremont, US
Huan Chen · Alpharetta, US
Qi Song · Beijing, CN

Key dates

Filing date	Oct 28, 2020
Grant date	May 18, 2021
Priority date	—
Expiry date	Oct 28, 2040

Classification

Technology area (CPC G)Physics
CPC primaryG06Q10/047
WIPO fieldMeasurement
WIPO sectorInstruments

Abstract

Methods and systems for generating a trained destination prediction model are provided. The method may include obtaining a plurality of historical orders corresponding to a plurality of users and determining a plurality of first features and a plurality of second features associated with the plurality of historical orders. The method may further include determining a plurality of transformed features based on the plurality of first features and a plurality of sets of cross features by correlating the plurality of second features. The method may further include obtaining a preliminary destination prediction model and training the preliminary destination prediction model to obtain a trained destination prediction model based on the plurality of transformed features and the plurality of sets of cross features.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.