# US 2021/0342649 A1 · Systems for Predicting a Terminal Event

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

- Number: US 2021/0342649 A1
- Document ID: US20210342649A1
- Status: Published application
- Kind: A1
- Application number: 16/866,261
- Filed: 2020-05-04
- Published: 2021-11-04
- Earliest claimed: 2020-05-04
- Assignees: Adobe Inc.
- Applicants: Adobe Inc.
- Inventors: Amit Doda; Gaurav Sinha; Kai Yeung Lau; Akangsha Sunil Bedmutha; Shiv Kumar Saini; Ritwik Sinha; Vaidyanathan Venkatraman; Niranjan Shivanand Kumbi; Omar Rahman; Atanu R. Sinha
- Web page: https://patentgrep.com/patent/US20210342649A1

## CPC

- G06F 3/04842: Selection of displayed objects or displayed text elements (G06F3/0482 takes precedence)
- G06F 3/0481: based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
- G06F 3/0482: Interaction with lists of selectable items, e.g. menus
- G06F 11/079: {Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36)}
- G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)}
- G06F 18/2113: by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
- G06F 18/2163: {Partitioning the feature space}
- G06F 18/241: relating to the classification model, e.g. parametric or non-parametric approaches
- G06F 18/2431: Multiple classes
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/045: Combinations of networks
- G06N 3/08: Learning methods
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

In implementations of systems for predicting a terminal event, a computing device implements a termination system to receive input data defining a period of time and a maximum event threshold. This system uses a classification model to generate event scores for a plurality of entity devices. Each of the event scores indicates a probability of an event occurrence for a corresponding entity device within a period of time. The plurality of entity devices are segmented into a first segment and a second segment based on an event score threshold. Entity devices included in the first segment have event scores greater than the event score threshold and entity devices included in the second segment have event scores below the event score threshold. The termination system generates an indication of a probability that a number of event occurrences for the entity devices included in the second segment exceeds the maximum even threshold within the period of time.

## Claims

### Claim 1

In a digital medium environment, a method implemented by a computing device, the method comprising:
receiving, by the computing device, a user input defining a period of time and a maximum event threshold;
generating, by the computing device using a classification model, event scores for a plurality of entity devices, each of the event scores indicating a probability of an event occurrence for a corresponding entity device of the plurality of entity devices within the period of time;
segmenting, by the computing device, the plurality of entity devices into a first segment and a second segment based on an event score threshold, entity devices included in the first segment having event scores above the event score threshold and entity devices included in the second segment having event scores below the event score threshold; and
generating, by the computing device, an indication of a probability that a number of event occurrences for the entity devices included in the second segment exceeds the maximum event threshold within the period of time, the indication generated for display in a user interface.

### Claim 2 (depends on 1)

The method as described in claim 1, wherein the classification model includes at least one of a logistic regression model, a random forest model, or an XGBoost model.

### Claim 3 (depends on 1)

The method as described in claim 1, further comprising forming communications for receipt by the entity devices included in the second segment.

### Claim 4 (depends on 3)

The method as described in claim 3, wherein the communications are e-mail communications of an entity-to-entity subscription service.

### Claim 5 (depends on 4)

The method as described in claim 4, wherein the event occurrences for the entity devices included in the second segment are terminations of the entity-to-entity subscription service.

### Claim 6 (depends on 1)

The method as described in claim 1, wherein the classification model is trained using training data describing prior interactions of training entity devices during a prior time period equal to the period of time.

### Claim 7 (depends on 6)

The method as described in claim 6, wherein the training data describes a positive class of the training entity devices for event occurrences during the prior time period and a negative class of the training entity devices for event non-occurrences during the prior time period.

### Claim 8 (depends on 7)

The method as described in claim 7, wherein the training data is generated by undersampling the negative class of the training entity devices, oversampling the positive class of the training entity devices, or using a weighting function.

### Claim 9 (depends on 1)

The method as described in claim 1, further comprising generating an indication of an average number of event occurrences for the entity devices included in the second segment during the period of time.

### Claim 10

A system comprising:
a termination module implemented at least partially in hardware of a computing device to:
receive interaction data describing prior interactions of a plurality of training entity devices;
determine a subset of data from the interaction data, the subset of data determined using time-stamps of the prior interactions and the subset of data describing prior interactions of the plurality of training entity devices within a prior time period equal to a period of time for predicting occurrences of events;
generate training data from the subset of data, the training data including indications of a recency and a frequency of the prior interactions within the prior period of time for each of the plurality of training entity devices;
train a classification model to predict the occurrences of events using the training data; and
generate an event score using the trained classification model, the event score indicating a probability of an event occurrence for an entity device within the period of time.

### Claim 11 (depends on 10)

The system as described in claim 10, wherein the classification model includes at least one of a logistic regression model, a random forest model, or an XGBoost model.

### Claim 12 (depends on 10)

The system as described in claim 10, wherein the termination module is further implemented to form a communication for receipt by the entity device.

### Claim 13 (depends on 12)

The system as described in claim 12, wherein the communication is an e-mail communication of an entity-to-entity subscription service.

### Claim 14 (depends on 13)

The system as described in claim 13, wherein the event occurrence is a termination of the entity-to-entity subscription service.

### Claim 15

One or more computer-readable storage media comprising instructions stored thereon that, responsive to execution by a computing device causes the computing device to perform operations including:
receiving a user input defining a period of time and a maximum event threshold;
generating event scores for a plurality of entity devices using a classification model, each of the event scores indicating a probability of an event occurrence for a corresponding entity device of the plurality of entity devices within the period of time;
segmenting the plurality of entity devices into a first segment and a second segment based on an event score threshold, entity devices included in the first segment having event scores above the event score threshold and entity devices included in the second segment having event scores below the event score threshold; and
generating, by the computing device, an indication of a probability that a number of event occurrences for the entity devices included in the second segment exceeds the maximum event threshold within the period of time, the indication generated for display in a user interface.

### Claim 16 (depends on 15)

The one or more computer-readable storage media as described in claim 15, wherein the classification model includes at least one of a logistic regression model, a random forest model, or an XGBoost model.

### Claim 17 (depends on 15)

The one or more computer-readable storage media as described in claim 15, the operations further including forming communications for receipt by the entity devices included in the second segment.

### Claim 18 (depends on 17)

The one or more computer-readable storage media as described in claim 17, wherein the communications are e-mail communications of an entity-to-entity subscription service.

### Claim 19 (depends on 18)

The one or more computer-readable storage media as described in claim 18, wherein the event occurrences for the entity devices included in the second segment are terminations of the entity-to-entity subscription service.

### Claim 20 (depends on 15)

The one or more computer-readable storage media as described in claim 15, wherein the classification model is trained using training data describing prior interactions of a plurality of training entity devices during a prior period of time equal to the period of time.
