# US 2026/0065237 A1 · Automated viral content detection

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

- Number: US 2026/0065237 A1
- Document ID: US20260065237A1
- Status: Published application
- Kind: A1
- Application number: 18/817,874
- Filed: 2024-08-28
- Published: 2026-03-05
- Earliest claimed: 2024-08-28
- Assignees: American Express Travel Related Services Company, Inc.
- Applicants: American Express Travel Related Services Company, Inc.
- Inventors: Jianqiu Chen; Bomnae Hur; Suraj Madnani; Ravi Varma
- Web page: https://patentgrep.com/patent/US20260065237A1

## CPC

- G06Q 10/40: Business processes related to social networking or social networking services
- G06N 20/00: Machine learning
- G06Q 30/0282: Rating or review of business operators or products
- 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
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A computing device can determine a negative sentiment score based at least in part on user posts to a social media platform within a predefined interval of time. Then, the computing device can compare the negative sentiment score to a moving average sentiment score to determine whether the negative sentiment score exceeds a threshold value based at least in part on the moving average sentiment score. The computing device can then generate an alert in response to a determination that the negative sentiment score exceeds the moving average sentiment score by the predefined amount. Finally, the alert can be formatted according to a recipient defined format and transmitted to a client device, thereby activating a client application to cause the client application to display the alert and enable the client device to connect to the computing device to view additional information about the alert.

## Claims

### Claim 1

A system, comprising:
a computing device comprising a processor and a memory; and
machine-readable instructions stored in the memory that, when executed by the processor, cause the computing device to at least:
train a machine learning model for sentiment analysis of a plurality of social media posts based at least in part on applying a machine learning algorithm to a training data set that comprises a sample text in a format of a respective social media platform, the machine learning model being trained to classify a respective social media post as at least one of positive, negative, or neutral;
identify an expiration of a predefined interval of time for obtaining user submitted content from a social media platform site;
preprocess the user submitted content by removing a media file from a plurality of user posts to generate a plurality of modified user posts in a plaintext format;
determine a negative sentiment score of a subset of posts with a negative sentiment based at least in part on the plurality of modified user posts, wherein the negative sentiment score is calculated by performing the sentiment analysis on the plurality of modified user posts to identify the subset of posts with the negative sentiment based at least in part on the machine learning model;
determine the negative sentiment score exceeds a first moving average sentiment score;
generate an alert in response to the determination that the negative sentiment score exceeds the first moving average sentiment score;
format the alert according to a defined format of a recipient;
transmit the alert to a client device associated with the recipient, wherein the alert activates a client application executing on the client device to cause the client application to display the alert and to enable the client device to connect to the computing device across a network to view additional information about the alert; and
halt the alert at the client device by instructing the client device to cease to display the alert on the client device based at least in part on a second moving average sentiment score that is less than a respective threshold value associated with the first moving average sentiment score.

### Claim 2 (depends on 1)

The system of claim 1, wherein the first moving average sentiment score is a moving average of the negative sentiment score for a defined preceding time period.

### Claim 3 (depends on 1)

The system of claim 1, wherein the machine-readable instructions that cause the computing device to determine the negative sentiment score further causes the computing device to at least:
for each post in the subset of posts with the negative sentiment, identify a number of reposts the post and a number of likes associated with the post; and
calculate a sum of the posts in the subset of posts with the negative sentiment and the number of reposts and the number of likes associated each post in the subset of posts with the negative sentiment, wherein the sum represents the negative sentiment score.

### Claim 4 (depends on 1)

The system of claim 1, wherein the machine-readable instructions that halt the alert at the client device by instructing the client device to cease to display the alert further cause the computing device to at least:
determine an alert time interval has passed since the transmission of the alert to the client device; and
compare the first moving average sentiment score to the second moving average sentiment score to determine whether the second moving average sentiment score is less than a threshold value that is based at least in part on the first moving average sentiment score.

### Claim 5 (depends on 1)

The system of claim 1, wherein the machine-readable instructions that halt the alert at the client device further cause the computing device to at least:
in response to the alert being triggered, calculate the second moving average sentiment score, the second moving average sentiment score representing a moving average of the negative sentiment score since the alert was initiated; and
compare the first moving average sentiment score to the second moving average sentiment score to determine whether the second moving average sentiment score is less than a threshold value that is based at least in part on the first moving average sentiment score.

### Claim 6 (depends on 1)

The system of claim 1, wherein the machine-readable instructions that halt the alert at the client device by instructing the client device to cease to display the alert further cause the computing device to at least:
recalculate the negative sentiment score for the subset of posts associated with the social media platform within the predefined interval of time;
compare the recalculated negative sentiment score to the first moving average sentiment score to determine whether the recalculated negative sentiment score is less than the threshold value, the threshold value being based at least in part on the first moving average sentiment score.

### Claim 7 (depends on 6)

The system of claim 6, wherein the alert is halted by further causing the computing device to at least:
transmit a subsequent email or a subsequent SMS message to the client device associated with the recipient.

### Claim 8

A computer-implemented method, comprising:
training, by at least one computing device, a machine learning model for sentiment analysis of a plurality of social media posts based at least in part on applying a machine learning algorithm to a training data set that comprises a sample text in a format of a respective social media platform, the machine learning model being trained to classify a respective social media post as at least one of positive, negative, or neutral;
removing, by the at least one computing device, a media file from a plurality of user posts to generate a plurality of modified user posts in a plaintext format;
calculating, by the at least one computing device, a negative sentiment score of a subset of posts with a negative sentiment based at least in part on the plurality of modified user posts, wherein the negative sentiment score is calculated by performing the sentiment analysis on the plurality of modified user posts to identify the subset of posts with the negative sentiment based at least in part on the machine learning model;
determining, by the at least one computing device, the negative sentiment score exceeds a first moving average sentiment score;
generating, by the at least one computing device, an alert in response to the determination that the negative sentiment score exceeds the first moving average sentiment score;
formatting, by the at least one computing device, the alert according to a defined format of a recipient; and
transmitting, by the at least one computing device, the alert to a client device associated with the recipient, wherein the alert activates a client application executing on the client device to cause the client application to display the alert and to enable the client device to connect to the at least one computing device across a network to view additional information about the alert; and
halting, by the at least one computing device, the alert at the client device by instructing the client device to cease to display the alert on the client device based at least in part on a second moving average sentiment score that is less than a respective threshold value associated with the first moving average sentiment score.

### Claim 9 (depends on 8)

The computer-implemented method of claim 8, wherein the first moving average sentiment score is a moving average of the negative sentiment score for a defined preceding time period.

### Claim 10 (depends on 8)

The computer-implemented method of claim 8, wherein calculating the negative sentiment score further comprises:
for each post in the subset of posts with the negative sentiment, identifying, by the at least one computing device, a number of reposts the post and a number of likes associated with the post; and
calculating, by the at least one computing device, a sum of the posts in the subset of posts with the negative sentiment and the number of reposts and the number of likes associated each post in the subset of posts with the negative sentiment, wherein the sum represents the negative sentiment score.

### Claim 11 (depends on 8)

The computer-implemented method of claim 8, wherein halting the alert at the client device by instructing the client device to cease to display the alert further comprises:
determining, by the at least one computing device, an alert time interval has passed since the transmission of the alert to the client device; and
comparing, by the at least one computing device, the first moving average sentiment score to a second moving average sentiment score to determine whether the second moving average sentiment score is less than a threshold value that is based at least in part on the first moving average sentiment score.

### Claim 12 (depends on 8)

The computer-implemented method of claim 8, wherein halting the alert at the client device by instructing the client device to cease to display the alert further comprises:
in response to the alert being triggered, calculating, by the at least one computing device, a second moving average sentiment score, the second moving average sentiment score representing a moving average of the negative sentiment score since the alert was initiated; and
comparing, by the at least one computing device, the first moving average sentiment score to the second moving average sentiment score to determine whether the second moving average sentiment score is less than a threshold value that is based at least in part on the first moving average sentiment score.

### Claim 13 (depends on 8)

The computer-implemented method of claim 8, wherein halting the alert at the client device by instructing the client device to cease to display the alert further comprises:
recalculating, by the at least one computing device, the negative sentiment score for the subset of posts associated with the social media platform within a predefined interval of time; and
comparing, by the at least one computing device, the recalculated negative sentiment score to the first moving average sentiment score to determine whether the recalculated negative sentiment score is less than the threshold value, the threshold value being based at least in part on the first moving average sentiment score.

### Claim 14 (depends on 13)

The computer-implemented method of claim 13, wherein the alert is halted by the method further comprising:
transmitting, by the at least one computing device, a subsequent email or a subsequent SMS message to the client device associated with the recipient.

### Claim 15

A non-transitory, computer-readable medium, comprising machine-readable instructions that, when executed by a processor of a computing device, cause the computing device to at least:
train a machine learning model for sentiment analysis of a plurality of social media posts based at least in part on applying a machine learning algorithm to a training data set that comprises a sample text in a format of a respective social media platform, the machine learning model being trained to classify a respective social media post as at least one of positive, negative, or neutral;
remove a media file from a plurality of user posts to generate a plurality of modified user posts in a plaintext format;
determine a negative sentiment score of a subset of posts with a negative sentiment based at least in part on the plurality of modified user posts, wherein the negative sentiment score is calculated by performing the sentiment analysis on the plurality of modified user posts to identify the subset of posts with the negative sentiment based at least in part on the machine learning model;
determine the negative sentiment score exceeds a first moving average sentiment score;
generate an alert in response to the determination that the negative sentiment score exceeds the first moving average sentiment score;
format the alert according to a defined format of a recipient;
transmit the alert to a client device associated with the recipient, wherein the alert activates a client application executing on the client device to cause the client application to display the alert and to enable the client device to connect to the computing device across a network to view additional information about the alert; and
halt the alert at the client device by instructing the client device to cease to display the alert on the client device based at least in part on a second moving average sentiment score that is less than a respective threshold value associated with the first moving average sentiment score.

### Claim 16 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the first moving average sentiment score is a moving average of the negative sentiment score for a defined preceding time period.

### Claim 17 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the machine-readable instructions that cause the computing device to determine the negative sentiment score further cause the computing device to at least:
for each post in the subset of posts with the negative sentiment, identify a number of reposts the post and a number of likes associated with the post; and
calculate a sum of the posts in the subset of posts with the negative sentiment and the number of reposts and the number of likes associated each post in the subset of posts with the negative sentiment, wherein the sum represents the negative sentiment score.

### Claim 18 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein and the machine-readable instructions that halt the alert at the client device by instructing the client device to cease to display the alert further cause the computing device to at least:
determine an alert time interval has passed since the transmission of the alert to the client device; and
compare the first moving average sentiment score to a second moving average sentiment score to determine whether the second moving average sentiment score is less than a threshold value that is based at least in part on the first moving average sentiment score.

### Claim 19 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the machine-readable instructions that halt the alert at the client device by instructing the client device to cease to display the alert further cause the computing device to at least:
in response to the alert being triggered, calculate a second moving average sentiment score, the second moving average sentiment score representing a moving average of the negative sentiment score since the alert was initiated; and
compare the first moving average sentiment score to the second moving average sentiment score to determine whether the second moving average sentiment score is less than a threshold value that is based at least in part on the first moving average sentiment score.

### Claim 20 (depends on 15)

The non-transitory, computer-readable medium of claim 15, wherein the machine-readable instructions that halt the alert at the client device by instructing the client device to cease to display the alert further cause the computing device to at least:
recalculate the negative sentiment score for the subset of posts associated with the social media platform within a predefined interval of time; and
compare the recalculated negative sentiment score to the first moving average sentiment score to determine whether the recalculated negative sentiment score is less than the threshold value, the threshold value being based at least in part on the first moving average sentiment score.
