# US 10,311,499 B1 · Clustering interactions for user missions

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 10,311,499 B1
- Document ID: US10311499B1
- Status: Granted
- Kind: B1
- Application number: 14/666,134
- Filed: 2015-03-23
- Granted: 2019-06-04
- Earliest claimed: 2015-03-23
- Assignees: Amazon Technologies, Inc.
- Applicants: Amazon Technologies, Inc.
- Inventors: Ilseo Kim; James Dai-fu Chan; Jean Joseph Tavernier
- Web page: https://patentgrep.com/patent/US10311499B1

## CPC

- G06Q 30/0631: {Recommending goods or services}
- G06F 16/2291: {User-Defined Types; Storage management thereof}
- G06Q 30/0251: Targeted advertisements
- G06Q 30/0255: {based on user history}
- G06Q 30/0601: Electronic shopping [e-shopping]
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Techniques for identifying clusters of user interactions and shopping missions may be provided. For example, the system may receive a history of interactions between a user and one or more network pages. The system may identify a most recent event from the history of interactions and identify a cluster that includes other events from the history of interactions that are of a same category as the most recent event. The determination of the cluster may be based in part on item attributes associated with the item presented on the at least one of the one or more network pages. The most recent event may then be associated with the cluster. In some examples, a shopping mission is determined and one or more notifications are provided to a user, merchant, or electronic marketplace in association with the identified shopping mission.

## Claims

### Claim 1

A computer-implemented method, comprising:
receiving, by a computing device, a history of network page interactions associated with a user and one or more network pages, the history of network page interactions including a time of each network page interaction in the history of network page interactions, and the history of network page interactions comprising an interaction that persists after the user closes a browser application at a user device;
identifying a most recent network-based event from the history of network page interactions based in part on the time of each network page interaction;
initiating, by the computing device, a clustering process to identify a cluster configured to include a set of network-based events from the history of network page interactions that are of a same category as the most recent network-based event;
identifying, by the computing device, an objective associated with the cluster based at least in part on the set of network-based events of the cluster;
determining, by the computing device, a current state of the objective utilizing a probabilistic model and the set of network-based events of the cluster, the probabilistic model being configured to identify the current state based at least in part on the set of network-based events of the cluster and historical network page interaction data associated with a plurality of users; and
executing operations to provide customized electronic content based at least in part on the current state.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the identified cluster corresponds with a description or title that describes interactions of the history of network page interactions associated with the cluster.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein items associated with the most recent network-based event and the set of network-based events in the identified cluster are substitutions for other items in the cluster.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining a similarity between the identified cluster and the most recent network-based event using a pairwise distance between the identified cluster and the most recent network-based event, wherein the determination of the similarity between the identified cluster and the most recent network-based event further considers at least one of user history, classification hierarchy, time of user network-based events, item attributes, or feedback from users associated with the identified cluster; and
associating the most recent network-based event with the identified cluster when the pairwise distance is less than a threshold.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, further comprising:
providing a notification to the user device that includes information associated with one or more items from the cluster, wherein the notification is provided by formatting a network page presented to the user device to include the notification based at least in part on the current state.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining that the objective is complete when the history of network page interactions associated with the user comprises an order for a first item associated with an item category that is also associated with a second item related to the most recent network-based event; and
when the objective is determined to be complete, providing a notification to the user device of the user that includes a recommended item from a different cluster.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the initiating of the clustering process begins after a search is submitted for an item in the same category as the most recent network-based event to trigger the clustering process.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein the initiating of the clustering process begins at a pre-determined time.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining that the current state indicates that the objective is incomplete when the history of network page interactions associated with the user indicates that an order has not been completed for a first item associated with an item category that is also associated with a second item related to the most recent network-based event, wherein the customized electronic content comprises a recommendation corresponding to an item related to the objective.

### Claim 10

One or more non-transitory computer-readable storage media collectively storing computer-executable instructions that, when executed by one or more computer systems, configure the one or more computer systems to collectively perform operations comprising:
receiving a history of network page interactions with one or more network pages, the history of network page interactions including a time of each network page interaction in the history of network page interactions, and the history of network page interactions comprising an interaction that persists after a user closes a browser application at a user device;
identifying a most recent network-based event from the history of network page interactions based in part on the time of each network page interaction;
initiating a clustering process to identify a cluster configured to include a set of network-based events from the history of network page interactions that are of a same category as the most recent network-based event;
identifying an objective associated with the cluster based at least in part on the set of network-based events of the cluster;
determining a current state of the objective utilizing a probabilistic model and the set of network-based events of the cluster, the probabilistic model being configured to identify the current state based at least in part on the set of network-based events of the cluster and historical network page interaction data associated with a plurality of users; and
executing operations to provide customized electronic content based at least in part on the current state.

### Claim 11 (depends on 10)

The one or more non-transitory computer-readable storage media of claim 10, wherein the identified cluster corresponds with a description or title that describes interactions of the history of network page interactions associated with the cluster.

### Claim 12 (depends on 10)

The one or more non-transitory computer-readable storage media of claim 10, wherein items associated with the most recent network-based event and the set of network-based events in the identified cluster are substitutions for other items in the cluster.

### Claim 13 (depends on 10)

The one or more non-transitory computer-readable storage media of claim 10, the operations further comprising:
determining a similarity between the identified cluster and the most recent network-based event using a pairwise distance between the identified cluster and the most recent network-based event; and
associating the most recent network-based event with the identified cluster when the pairwise distance is less than a threshold, wherein the determination of the similarity between the identified cluster and the most recent network-based event further considers at least one of user history, classification hierarchy, time of user network-based events, item attributes, or feedback from users associated with the identified cluster.

### Claim 14 (depends on 10)

The one or more non-transitory computer-readable storage media of claim 10, the operations further comprising:
providing a notification to the user device that includes information associated with one or more items from the cluster.

### Claim 15 (depends on 10)

The one or more non-transitory computer-readable storage media of claim 10, the operations further comprising:
identifying that the history of network page interactions comprises an order of an item for the same category as the most recent network-based event, wherein the objective is identified as complete when the history of network page interactions comprises the order; and
when the objective is complete, providing a notification to the user device that includes a recommended item from a different cluster.

### Claim 16

A computing device comprising:
a memory configured to store computer-executable instructions; and
a processor in communication with the memory configured to execute the computer-executable instructions to at least:
receive a history of network page interactions with one or more network pages, the history of network page interactions including a time of each network page interaction in the history of network page interactions, and the history of network page interactions comprising an interaction that persists after a user closes a browser application at a user device;
identify a most recent network-based event from the history of network page interactions based in part on the time of each network page interaction;
initiate a clustering process to identify a cluster configured to include a set of network-based events from the history of network page interactions that are of a same category as the most recent network-based event;
identify an objective associated with the cluster based at least in part on the set of network-based events of the cluster;
determine a current state of the objective utilizing a probabilistic model and the set of network-based events of the cluster, the probabilistic model being configured to identify the current state based at least in part on the set of network-based events of the cluster and historical network page interaction data associated with a plurality of users; and
execute operations to provide customized electronic content based at least in part on the current state.

### Claim 17 (depends on 16)

The computing device of claim 16, wherein the identified cluster corresponds with a description or title that describes interactions of the history of network page interactions associated with the cluster.

### Claim 18 (depends on 16)

The computing device of claim 16, wherein items associated with the most recent network-based event and the set of network-based events in the identified cluster are substitutions for other items in the cluster.

### Claim 19 (depends on 16)

The computing device of claim 16, the instructions further configured to:
provide a notification that includes information associated with one or more items from the cluster, wherein the notification is provided by formatting a network page to include the notification based at least in part on the current state.

### Claim 20 (depends on 16)

The computing device of claim 16, the instructions further configured to:
determine that the objective associated with the cluster is complete when the history of network page interactions comprises an order for a first item associated with an item category that is also associated with a second item related to the most recent network-based event; and
when the objective is determined to be complete, provide a notification to the user device that includes a recommended item from a different cluster.