# US 2024/0202095 A1 · Content item selection and click probability determination based upon accidental click events

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

- Number: US 2024/0202095 A1
- Document ID: US20240202095A1
- Status: Published application
- Kind: A1
- Application number: 18/407,022
- Filed: 2024-01-08
- Published: 2024-06-20
- Earliest claimed: 2021-06-14
- Assignees: Yahoo Ad Tech LLC
- Applicants: Yahoo Ad Tech LLC
- Inventors: Naama Haramaty-Krasne; Yohay Kaplan; Oren Shlomo Somekh; Alexander Shtoff
- Web page: https://patentgrep.com/patent/US20240202095A1

## CPC

- G06F 3/0482: Interaction with lists of selectable items, e.g. menus
- G06F 11/3438: {monitoring of user actions (tracking the activity of the user H04L67/535)}
- G06F 3/04842: Selection of displayed objects or displayed text elements (G06F3/0482 takes precedence)
- G06F 16/95: Retrieval from the web
- G06N 20/00: Machine learning
- G06Q 30/0248: {Avoiding fraud}
- G06Q 30/0251: Targeted advertisements
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- 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

## Abstract

In an example, sets of event information associated with events may be identified. The events may include intentional click events, accidental click events and/or skip events. Accidental click probabilities associated with the accidental click events and/or the skip events may be determined. Machine learning model training may be performed, using the sets of event information associated with the events and labels associated with the events, to generate a first machine learning model. The labels may include second labels associated with the intentional click events and/or third labels associated with the accidental click events and/or the skip events. The second labels may correspond to an intentional click classification. The third labels may be based upon the accidental click probabilities. Click probabilities associated with content items may be determined using the first machine learning model. A content item may be selected for presentation via a client device based upon the click probabilities.

## Claims

### Claim 1

A method, comprising:
identifying a first plurality of sets of event information associated with a first plurality of events, wherein the first plurality of sets of event information comprises:
a second plurality of sets of event information associated with a plurality of accidental click events of the first plurality of events; and
a third plurality of sets of event information associated with a plurality of skip events of the first plurality of events;
determining a plurality of accidental click probabilities associated with a second plurality of events comprising the plurality of accidental click events and the plurality of skip events, wherein the determining the plurality of accidental click probabilities comprises:
determining a first accidental click probability, associated with a first accidental click event of the plurality of accidental click events, based upon a first set of event information associated with the first accidental click event; and
determining a second accidental click probability, associated with a first skip event of the plurality of skip events, based upon a second set of event information associated with the first skip event; and
performing machine learning model training, using the first plurality of sets of event information associated with the first plurality of events and a first plurality of labels associated with the first plurality of events, to generate a first machine learning model.

### Claim 2 (depends on 1)

The method of claim 1, wherein:
the first plurality of labels comprises a second plurality of labels associated with the second plurality of events; and
labels of the second plurality of labels are based upon the plurality of accidental click probabilities and comprise:
a first label, associated with the first accidental click event, based upon the first accidental click probability; and
a second label, associated with the first skip event, based upon the second accidental click probability.

### Claim 3 (depends on 1)

The method of claim 1, wherein:
the determining the plurality of accidental click probabilities is performed using a second machine learning model.

### Claim 4 (depends on 3)

The method of claim 3, comprising:
performing machine learning model training, using a fourth plurality of sets of event information associated with a second plurality of accidental click events and a fifth plurality of sets of event information associated with a second plurality of skip events, to generate the second machine learning model.

### Claim 5 (depends on 1)

The method of claim 1, comprising:
determining a first click probability, using the first machine learning model, based upon at least one of client information associated with a client device, content item information associated with a first content item or internet resource information associated with an internet resource associated with a request for content.

### Claim 6 (depends on 1)

The method of claim 1, comprising:
presenting a second content item, via a client device, on a first internet resource, wherein the second content item is selected via the client device during the presenting the second content item; and
detecting occurrence of the first accidental click event based upon a determination that a dwell time associated with the selection of the second content item is less than a threshold dwell time, wherein the dwell time comprises a time during which a second internet resource, accessed by the client device responsive to the selection of the second content item, is presented via the client device.

### Claim 7 (depends on 6)

The method of claim 6, comprising:
determining the dwell time based upon a first time and a second time, wherein:
the first time corresponds to at least one of:
a time at which the second content item is selected via the client device; or
a time at which the second internet resource is accessed responsive to the selection of the second content item; and
the second time corresponds to at least one of:
a time at which the client device returns to the first internet resource after the second content item is selected; or
a time at which the client device leaves the second internet resource.

### Claim 8 (depends on 1)

The method of claim 1, comprising:
presenting a second content item, via a client device, on a first internet resource; and
detecting occurrence of the first skip event based upon a determination that the second content item is not selected via the client device during the presenting the second content item.

### Claim 9 (depends on 8)

The method of claim 8, wherein:
the second accidental click probability is representative of a probability that presentation of the second content item via the client device would be followed by occurrence of an accidental click event.

### Claim 10 (depends on 1)

The method of claim 1, comprising:
presenting a second content item, via a client device, on a first internet resource, wherein the second content item is selected via the client device during the presenting the second content item; and
detecting occurrence of a first intentional click event of the first plurality of events based upon a determination that a dwell time associated with the selection of the second content item exceeds a threshold dwell time, wherein the dwell time comprises a time during which a second internet resource, accessed by the client device responsive to the selection of the second content item, is presented via the client device.

### Claim 11 (depends on 5)

The method of claim 5, wherein:
the determining the first click probability using the first machine learning model is performed based upon one or more features associated with the request for content.

### Claim 12 (depends on 5)

The method of claim 5, wherein:
the first click probability is representative of a probability of receiving a selection of the first content item responsive to presenting the first content item via the client device.

### Claim 13

A computing device comprising:
a processor; and
memory comprising processor-executable instructions that when executed by the processor cause performance of operations, the operations comprising:
identifying a first plurality of sets of event information associated with a first plurality of events, wherein the first plurality of sets of event information comprises:
a second plurality of sets of event information associated with a plurality of skip events of the first plurality of events;
determining a plurality of accidental click probabilities associated with the plurality of skip events, wherein the determining the plurality of accidental click probabilities comprises determining a first accidental click probability, associated with a first skip event of the plurality of skip events, based upon a first set of event information associated with the first skip event; and
performing machine learning model training, using the first plurality of sets of event information associated with the first plurality of events and a first plurality of labels associated with the first plurality of events, to generate a first machine learning model.

### Claim 14 (depends on 13)

The computing device of claim 13, wherein:
the first plurality of labels comprises a second plurality of labels associated with the plurality of skip events; and
labels of the second plurality of labels are based upon the plurality of accidental click probabilities and comprise a first label, associated with the first skip event, based upon the first accidental click probability.

### Claim 15 (depends on 13)

The computing device of claim 13, the operations comprising:
performing machine learning model training, using a third plurality of sets of event information associated with a plurality of accidental click events and a fourth plurality of sets of event information associated with a second plurality of skip events, to generate a second machine learning model, wherein the determining the plurality of accidental click probabilities is performed using the second machine learning model.

### Claim 16 (depends on 15)

The computing device of claim 15, the operations comprising:
presenting a second content item, via a client device, on a first internet resource, wherein the second content item is selected via the client device during the presenting the second content item; and
detecting occurrence of a first accidental click event of the plurality of accidental click events based upon a determination that a dwell time associated with the selection of the second content item is less than a threshold dwell time, wherein the dwell time comprises a time during which a second internet resource, accessed by the client device responsive to the selection of the second content item, is presented via the client device.

### Claim 17 (depends on 13)

The computing device of claim 13, the operations comprising:
determining a first click probability, using the first machine learning model, based upon at least one of client information, content item information or internet resource information.

### Claim 18 (depends on 13)

The computing device of claim 13, the operations comprising:
presenting a second content item, via a client device, on a first internet resource, wherein the second content item is selected via the client device during the presenting the second content item; and
detecting occurrence of a first intentional click event of the first plurality of events based upon a determination that a dwell time associated with the selection of the second content item exceeds a threshold dwell time, wherein the dwell time comprises a time during which a second internet resource, accessed by the client device responsive to the selection of the second content item, is presented via the client device.

### Claim 19

A non-transitory machine readable medium having stored thereon processor-executable instructions that when executed cause performance of operations, the operations comprising:
identifying a first plurality of sets of event information associated with a first plurality of events, wherein the first plurality of sets of event information comprises:
a second plurality of sets of event information associated with a plurality of accidental click events of the first plurality of events;
determining a plurality of accidental click probabilities associated with the plurality of accidental click events, wherein the determining the plurality of accidental click probabilities comprises determining a first accidental click probability, associated with a first accidental click event of the plurality of accidental click events, based upon a first set of event information associated with the first accidental click event; and
performing machine learning model training, using the first plurality of sets of event information associated with the first plurality of events and a first plurality of labels associated with the first plurality of events, to generate a first machine learning model.

### Claim 20 (depends on 19)

The non-transitory machine readable medium of claim 19, wherein:
the first plurality of labels comprises a second plurality of labels associated with the plurality of accidental click events; and
labels of the second plurality of labels are based upon the plurality of accidental click probabilities and comprise a first label, associated with the first accidental click event, based upon the first accidental click probability.
