| CPC H04L 67/535 (2022.05) [G06F 9/453 (2018.02); G06F 16/955 (2019.01); G06N 20/00 (2019.01); H04L 67/306 (2013.01)] | 20 Claims |
|
1. A method, comprising:
using, by a computer system in an online payment service, a machine learning model to analyze online actions taken by a particular user associated with a particular account of the online payment service;
receiving, by the computer system from a user device, an access request associated with the particular account; and
in response to the access request:
predicting, by the computer system, a next action to be taken by the particular user, wherein the predicting includes:
based on the analyzed online actions of the particular user, updating one or more profile attributes associated with the particular account;
determining priority scores for a plurality of potential next actions based on the one or more updated profile attributes; and
selecting, based on the determined priority scores, at least one of the potential next actions as a predicted next action;
sending, by the computer system, the predicted next action to the user device;
in response to an indication of a selection of a next action by the particular user, predicting, by the computer system based on the selection of the next action, a subsequent action to be taken by the particular user; and
sending, by the computer system, the predicted subsequent action to the user device.
|