Identifying the end of an on-line cart session

• US 10,185,987 B2
• Filed: 07/24/2015
• Issued: 01/22/2019
• Est. Priority Date: 07/24/2015
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Inc.
• Inventors: Jhamtani, Harsh, Revankar, Shriram V. S., Sinha, Moumita, Srinivasan, Balaji Vasan, Natarajan, Anandhavelu
• Primary Examiner(s): Miller, Brittney N
• Application Number:
  US14/808,883
• Publication Number:
  US 20170024807A1
• Time in Patent Office: 1,278 Days
• Field of Search: 705 268, 705 271
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

In embodiments of identifying the end of an on-line cart session, an analytics application captures user click inputs on pages of a Web site, where the user click inputs include adding one or more items for purchase to an on-line cart associated with the Web site. The analytics application then utilizes a predictive model, as well as user and session features of the on-line cart session, to predict whether a previous user click input is the last user click input associated with the on-line cart session, indicating an end of the session. A notification can then be provided that the on-line cart session has ended based on the prediction of the last user click input associated with the on-line cart session. The analytics application or the marketer can then retarget a user associated with the on-line cart session, such as with a message pertaining to the on-line cart session.

Claims

1. A method implemented to identify when a current on-line cart session associated with a user has ended and provide an indication of when to retarget the user, the method comprising:

capturing user click inputs on one or more pages of a website, the user click inputs captured by an analytics application on a server device during the current on-line cart session, the user click inputs including adding an item for purchase to an on-line cart associated with the website;

predicting, by the analytics application using a predictive model with a trained classifier, whether a previous user click input is a last user click input associated with the current on-line cart session, the last user click input indicating an end of the current on-line cart session, the predicting based at least in part on a duration of inactivity during the current on-line cart session, the predictive model developed from a data set of past on-line cart sessions and user website interactions of multiple previous users and by determining a statistical distribution from the data set, wherein the statistical distribution is based on a time duration between successive user click inputs and a time duration after the last user click input until a defined end of a past on-line cart session, wherein the trained classifier is trained based at least in part on the statistical distribution; and

in response to the predicting that one of the user click inputs is the last user click input, providing a notification that the current on-line cart session has ended.

2. The method as recited in claim 1, further comprising:

retargeting the user associated with the user click inputs and the current on-line cart session in response to said predicting the last user click input associated with the current on-line cart session.

3. The method as recited in claim 2, wherein said retargeting the user comprises communicating a message pertaining to the current on-line cart session to a computing device associated with the user.

4. The method as recited in claim 1, wherein said predicting the last user click input associated with the current on-line cart session is based on the predictive model, and is based on user features and session features of the current on-line cart session.

5. The method as recited in claim 4, wherein the user features and session features of the current on-line cart session comprise one or more of:

a number of the user click inputs in the current on-line cart session;

a time duration of the current on-line cart session;

a time difference between the user click inputs; and

a total number of items added to the on-line cart associated with the website.

6. The method as recited in claim 4, wherein the session features of the predictive model include at least:

a time duration since a last purchase, a time duration since a previous on-line cart session, and a mean session length of the past on-line cart sessions.

7. The method as recited in claim 1, wherein said providing the notification that the current on-line cart session has ended includes the analytics application initiating the notification to a marketer associated with the website that the current on-line cart session associated with the user has ended.

8. The method as recited in claim 1, wherein the predictive model is developed based on the multiple previous users as a representation of any subsequent user that initiates an on-line cart session.

9. The method as recited in claim 1, wherein said developing the predictive model comprises:

determining user features and session features of the past on-line cart sessions and the user website interactions; and

training the trained classifier based on the statistical distribution, the user features, and the session features of the past on-line cart sessions and the user website interactions.

10. A device implemented to identify when a current on-line cart session associated with a user has ended and provide an indication of when to retarget the user, the device comprising:

a memory configured to maintain user click inputs captured from one or more pages of a website and maintain a data set of past on-line cart sessions and user website interactions of multiple previous users, the user click inputs including adding an item for purchase to an on-line cart associated with the website;

a processor to implement an analytics application that is configured to;

predict, using a predictive model with a trained classifier, whether a previous user click input is a last user click input associated with the current on-line cart session, the last user click input indicating an end of the current on-line cart session, the prediction based at least in part on a duration of inactivity during the current on-line cart session, the predictive model developed by determining a statistical distribution from the data set, wherein the statistical distribution is based on a time duration between successive user click inputs and a time duration after the last user click input until a defined end of a past on-line cart session, wherein the trained classifier is trained based at least in part on the statistical distribution; and

in response to the prediction that one of the user click inputs is the last user click input, provide a notification that the current on-line cart session has ended.

11. The device as recited in claim 10, wherein the analytics application is configured to initiate retargeting the user associated with the user click inputs and the current on-line cart session in response to the prediction of the last user click input associated with the current on-line cart session.

12. The device as recited in claim 10, wherein the analytics application is configured to predict the last user click input associated with the current on-line cart session based on the predictive model, and based on user features and session features of the current on line cart session.

13. The device as recited in claim 12, wherein the user features and the session features of the current on-line cart session comprise one or more of:

a number of the user click inputs in the current on-line cart session;

a time duration of the current on-line cart session;

a time difference between the user click inputs; and

a total number of items added to the on-line cart associated with the website.

14. The device as recited in claim 12, wherein the analytics application is configured to initiate the notification to a marketer associated with the website that the current on-line cart session associated with the user has ended.

15. The device as recited in claim 10, wherein the predictive model is developed based on the multiple previous users as a representation of any subsequent user that initiates an on-line cart session.

16. The device as recited in claim 10, wherein to develop the predictive model, the analytics application is configured to:

determine user features and session features of the past on-line cart sessions and the user website interactions; and

train the trained classifier based on the statistical distribution, the user features, and the session features of the past on-line cart sessions and the user website interactions.

17. A method implemented by an analytics application on a server device to develop a predictive model that is utilized to identify when a current on-line cart session associated with a user has ended and provide an indication of when to retarget the user, the method implemented by the analytics application comprising:

obtaining a data set of past on-line cart sessions and user website interactions of multiple previous users;

determining a statistical distribution from the data set of the past on-line cart sessions and the user website interactions, the statistical distribution based on a time duration between successive user click inputs and a time duration after a last user click input until a defined end of a past on-line cart session;

determining user features and session features of the past on-line cart sessions and the user website interactions;

training a classifier based on the statistical distribution, the user features, and the session features of the past on-line cart session and the user website interactions;

capturing user click inputs on one or more pages of a website during the current on-line cart session, the user click inputs including adding an item for purchase to an on-line cart associated with the website;

utilizing the predictive model to predict whether a previous user click input is the last user click input associated with the current on-line cart session, indicating an end of the current on-line cart session; and

in response to the prediction that one of the user click inputs is the last user click input, providing a notification that the current on-line cart session has ended.

18. The method as recited in claim 17, further comprising:

developing the predictive model based on the multiple previous users as a representation of any subsequent user that initiates an on-line cart session.

19. The method as recited in claim 17, wherein said providing the notification that the current on-line cart session has ended includes initiating the notification to a marketer associated with the website that the current on-line cart session associated with the user has ended.

20. The method as recited in claim 17, wherein the session features of the predictive model include at least:

a time duration since a last purchase, a time duration since a previous on-line cart session, and a mean session length of the past on-line cart sessions.