Behavioral prediction for targeted end users

Patent metadata

Field	Value
Publication number	US-11301761-B2
Application number	US-201916241456-A
Country	US
Kind code	B2
Filing date	Jan 7, 2019
Priority date	Aug 17, 2015
Publication date	Apr 12, 2022
Grant date	Apr 12, 2022

Abstract

Behavioral prediction for targeted end users is described. In one or more example embodiments, a computer-readable storage medium has multiple instructions that cause one or more processors to perform multiple operations. Targeted selectstream data is obtained from one or more indications of data object requests corresponding to a targeted end user. A targeted directed graph is constructed based on the targeted selectstream data. A targeted graph feature vector is computed based on one or more invariant features associated with the targeted directed graph. A behavioral prediction is produced for the targeted end user by applying a prediction model to the targeted graph feature vector. In one or more example embodiments, the prediction model is generated based on multiple graph feature vectors respectively corresponding to multiple end users. In one or more example embodiments, a tailored opportunity is determined responsive to the behavioral prediction and issued to the targeted end user.

First claim

What is claimed is:

1. A computing device implemented to predict targeted end user behavior in a digital medium environment, the computing device comprising:
a memory to maintain selectstream data indicative of data objects interaction by multiple end users in a computing environment, and the memory to maintain distinct classes that correspond to different behavioral traits of the multiple end users who are classified into one or more clusters based on the different behavioral traits derived from the selectstream data;
a processor system to implement a behavioral prediction engine at least partially in hardware of the computing device, the behavioral prediction engine configured to:
construct a directed graph based on the selectstream data corresponding to a new end user, the directed graph including one or more invariant features that translate to marketing analytics representative of the data objects interaction by the new end user in the computing environment;
determine isomorphic directed graphs based on a correlation between the directed graph of the new end user and directed graphs corresponding to one or more of the multiple end users, the isomorphic directed graphs including a same set of the one or more invariant features that represent the data objects interaction by the new end user and the one or more multiple end users in the computing environment;
predict a future behavior of the new end user using a machine learning system applied to the isomorphic directed graphs, the future behavior predicted based on the correlation of the isomorphic directed graphs as related to the distinct classes that correspond to different behavioral traits of the multiple end users, a directed opportunity tailored to the new end user based on the predicted future behavior of the new end user as derived from the selectstream data; and
classify the new end user for membership into a selected cluster of the one or more clusters based on the different behavioral trait associated with the selected cluster as known from the multiple end users who are classified in the selected cluster.

2. The device as described in claim 1 , wherein each isomorphic directed graph is representative of a path of travel as a corresponding end user traverses multiple ones of the data objects.

3. The device as described in claim 1 , wherein the behavioral prediction engine includes a directed graph construction module configured to:
assign vertices of a given isomorphic directed graph based on the data objects interactions by a corresponding end user; and
define respective directed edges of the given isomorphic directed graph based on respective pairs of source data objects and destination data objects that are traversed by the corresponding end user, the pairs of source data objects and destination data objects identified in the selectstream data of the corresponding end user.

4. The device as described in claim 1 , wherein the one or more invariant features comprise at least one intuitive invariant feature that translates to a physical marketing concept and at least one abstract invariant feature that pertains to isomorphic directed graph topology.

5. The device as described in claim 1 , wherein the behavioral prediction engine includes a graph feature vector computation module configured to compute multiple respective real values that correspond to multiple respective invariant features of the one or more invariant features for each isomorphic directed graph.

6. The device as described in claim 1 , wherein the behavioral prediction engine includes a prediction model generation module configured to perform a clustering operation to separate multiple graph feature vectors into the one or more clusters based on one or more similarities between or among different ones of the multiple graph feature vectors.

7. The device as described in claim 1 , wherein the behavioral prediction engine includes a prediction model generation module configured to perform a training operation on a classifier using multiple graph feature vectors as a training set of graph feature vectors associated with a classification category.

8. The device as described in claim 1 , wherein the behavioral prediction engine is configured to apply a prediction model to a graph feature vector of an isomorphic directed graph that is derived from the selectstream data to determine the behavioral prediction of the future behavior for the new end user.

9. A system implemented to predict targeted end user behavior to facilitate opportunity tailoring in a digital medium environment, the system comprising:
one or more computing devices that implement a behavioral prediction engine at least partially in hardware, the behavioral prediction engine configured to generate a prediction model from an input of selectstream data performing operations comprising:
obtaining the selectstream data from one or more indications of data object requests corresponding to a targeted end user, the selectstream data indicative of the data object requests by the targeted end user in a computing environment;
constructing a directed graph based on the selectstream data, the directed graph including one or more invariant features that translate to marketing analytics reflective of the data object requests by the targeted end user in the computing environment;
determining isomorphic directed graphs based on a correlation between the directed graph of the targeted end user and directed graphs corresponding to one or more multiple end users, the isomorphic directed graphs including a same set of the one or more invariant features that represent the data object requests by the targeted end user and the one or more multiple end users in the computing environment;
predicting a future behavior of the targeted end user by applying the prediction model to correlate the isomorphic directed graphs based on different behavioral traits of the multiple end users, a directed opportunity tailored to the targeted end user based on the predicted future behavior of the targeted end user from the isomorphic directed graphs as derived from the selectstream data; and
classifying the targeted end user in one or more classification categories, each associated with a particular behavioral attribute as determined from the multiple end users.

10. The system as described in claim 9 , wherein the data object requests comprise requests for website pages that are each identified by a uniform resource locator (URL), and the obtaining comprises receiving the data object requests for the website pages from an end-user device of the targeted end user.

11. The system as described in claim 9 , wherein the future behavior prediction comprises at least one of: an indication that the targeted end user is likely to make a purchase, an indication that the targeted end user is likely to join a customer associative program, or an indication that the targeted end user is likely to abandon an electronic cart without completing a purchase.

12. The system as described in claim 9 , wherein the predicting the future behavior comprises:
identifying one or more graph feature vectors based on a similarity to a feature vector of the directed graph, the one or more graph feature vectors corresponding to one or more of the multiple end users that exhibited the particular behavioral attribute; and
assigning to the future behavior prediction the particular behavioral attribute.

13. The system as described in claim 12 , wherein a measure of the similarity comprises an L-p norm distance, and the identifying comprises determining the L-p norm distance between the feature vector of the directed graph and the one or mo

Assignees

• Adobe Inc

Inventors

• Krishnamurthy Balaji
• Singla Tushar

Classifications

• G06F16/9024

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q30/0241

  Advertisements · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06Q30/00

  Commerce · CPC title