Predicting and preventing negative user experience in a network service

Patent metadata

Field	Value
Publication number	US-12332903-B2
Application number	US-202318231993-A
Country	US
Kind code	B2
Filing date	Aug 9, 2023
Priority date	Jan 31, 2019
Publication date	Jun 17, 2025
Grant date	Jun 17, 2025

Abstract

A system can monitor event data corresponding to a current user experience of a requesting user during a current application session with a network service. Based on the event data, the system generates one or more representations corresponding to the current user experience of the requesting user, and executes a machine learning model to process the one or more representations in order to predict a negative user experience for the requesting user within a future time frame during the current application session. In response to predicting the negative user experience, the system implements one or more corrective actions during the current application session through the service application to prevent or mitigate the predicted negative user experience.

First claim

What is claimed is:

1. A computing system implementing a network service, comprising:
a network communication interface to communicate, over one or more networks, with a service application executing on computing devices of requesting users of the network service;
one or more processors; and
one or more memory resources storing instructions that, when executed by the one or more processors, cause the computing system to:
monitor, over the one or more networks, event data corresponding to a current user experience of a requesting user during a current application session with the network service, the requesting user being matched to a matched service provider;
based on the event data, generate a first representation corresponding to the current user experience of the requesting user and a second representation corresponding to a general behavior of the matched service provider, the first representative further corresponding to historical utilization of the network service by the requesting user, and the second representation further corresponding to route information of the matched service provider;
execute a machine learning model to dynamically process the second representation to determine, based on the route information and the general behavior of the matched service provider, that the matched service provider is attempting to induce cancelation of a match between the matched service provider and the requesting user;
execute the machine learning model to dynamically process the first representation to predict a negative user experience for the requesting user within a future time frame during the current application session, the predicted negative user experience corresponding to a prediction that the requesting user will cancel the match between the matched service provider and the requesting user based, at least in part, on the matched service provider attempting to induce cancelation; and
in response to predicting the negative user experience, implement one or more corrective actions during the current application session through the service application to prevent or mitigate the predicted negative user experience.

2. The computing system of claim 1 , wherein the predicted negative user experience further comprises a prediction that the requesting user will abandon the network service.

3. The computing system of claim 1 , wherein the event data further comprises location data indicating a dynamic location of the matched service provider matched to rendezvous with the requesting user to fulfill the service request configured by the requesting user.

4. The computing system of claim 1 , wherein the one or more corrective actions comprise at least one of transmitting a notification to the computing device of the matched service provider, providing a service benefit to the requesting user, inputting a demerit in a provider profile of the matched service provider, or automatically matching the requesting user with a different service provider.

5. The computing system of claim 1 , wherein the executed instructions further cause the computing system to:
determine, based on historical user data indicating the historical utilization of the network service by the requesting user, an engagement level of the requesting user with the network service.

6. The computing system of claim 5 , wherein the executed instructions cause the computing system to determine a set of corrective actions based, at least in part, on the engagement level of the requesting user.

7. The computing system of claim 1 , wherein the network service comprises an on-demand transport service, and wherein the matched service provider comprises a driver of the on-demand transport service.

8. The computing system of claim 7 , wherein the on-demand transport service includes at least one of transporting the requesting user to a destination inputted in the service request, or delivering a requested item to the requesting user.

9. The computing system of claim 1 , wherein the event data comprises at least one of location data, input data on the service application by the requesting user, or sensor data from a computing device of the requesting user.

10. A non-transitory computer readable medium storing instructions that, when executed by one or more processors of a computing system, cause the computing system to:
communicate, over one or more networks, with a service application executing on computing devices of requesting users of a network service;
monitor, over the one or more networks, event data corresponding to a current user experience of a requesting user during a current application session with the network service, the requesting user being matched to a matched service provider;
based on the event data, generate a first representation corresponding to the current user experience of the requesting user and a second representation corresponding to a general behavior of the matched service provider, the first representation further corresponding to historical utilization of the network service by the requesting user, and the second representation further corresponding to route information of the matched service provider;
execute a machine learning model to dynamically process the second representation to determine, based on the route information and the general behavior of the matched service provider, that the matched service provider is attempting to induce cancelation of a match between the matched service provider and the requesting user;
execute the machine learning model to dynamically process the first representation to predict a negative user experience for the requesting user within a future time frame during the current application session, the predicted negative user experience corresponding to a prediction that the requesting user will cancel the match between the matched service provider and the requesting user based, at least in part, on the matched service provider attempting to induce cancelation; and
in response to predicting the negative user experience, implement one or more corrective actions during the current application session through the service application to prevent or mitigate the predicted negative user experience.

11. The non-transitory computer readable medium of claim 10 , wherein the predicted negative user experience further comprises a prediction that the requesting user will abandon the network service.

12. The non-transitory computer readable medium of claim 10 , wherein the event data further comprises location data indicating a dynamic location of the matched service provider matched to rendezvous with the requesting user to fulfill the service request configured by the requesting user.

13. The non-transitory computer readable medium of claim 10 , wherein the one or more corrective actions comprise at least one of transmitting a notification to the computing device of the matched service provider, providing a service benefit to the requesting user, inputting a demerit in a provider profile of the matched service provider, or automatically matching the requesting user with a different service provider.

14. The non-transitory computer readable medium of claim 10 , wherein the executed instructions further cause the computing system to:
determine, based on historical user data indicating the historical utilization of the network service by the requesting user, an engagement level of the requesting user with the network service.

15. The non-transitory computer readable medium of claim 14 , wherein the executed instructions cause the computing system to determine a set of corrective actions based, at least in part

Assignees

• Uber Technologies Inc

Inventors

• Goldstein Lawrence Benjamin
• Vora Arjun
• Malkani Jai
• Joshi Nikhil
• Walther Eckart
• Namazifar Mahdi
• Brown Kathy
• Kannan Praveen
• Bodapati Arun
• Bell Fran

Classifications

• G06Q50/40 Primary

  Business processes related to the transportation industry (shipping G06Q10/083) · CPC title

• H04L67/535

  Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34) · CPC title

• G06N20/00

  Machine learning · CPC title

• H04W4/023

  using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds · CPC title

• H04W4/40

  for vehicles, e.g. vehicle-to-pedestrians [V2P] · CPC title