# US 10,120,859 B2 · Message sentiment analyzer and message preclusion

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

- Number: US 10,120,859 B2
- Document ID: US10120859B2
- Status: Granted
- Kind: B2
- Application number: 15/005,132
- Filed: 2016-01-25
- Published: 2016-05-26
- Granted: 2018-11-06
- Earliest claimed: 2013-12-16
- Assignees: FairWords, Inc.
- Applicants: FairWords, Inc.
- Inventors: Anish Parikh; Evan M. Caron; Vadim Polosatov
- Web page: https://patentgrep.com/patent/US10120859B2

## CPC

- G06F 40/274: Converting codes to words; Guess-ahead of partial word inputs
- H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
- H04L 67/30: Profiles
- H04L 51/04: Real-time or near real-time messaging, e.g. instant messaging [IM]
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

This disclosure describes systems, methods, and apparatus for generating a computer response to an incoming message. Incoming messages can be analyzed to determine a sentiment, and this sentiment can be put to various uses. For instance, when a reply message is created, the sentiment can be used to make suggestions or prompt the message creator to use different language tailored to the sentiment of the original sender. In other cases, creation of the message may be precluded until existing text is modified to better suit the recipient's sentiment.

## Claims

### Claim 1

A method of generating computer responses to external communications, the method carried out via a processing device, the method comprising:
identifying one or more sequences of characters in an incoming electronic message entering a private network, the sequence representing an idea;
parsing the characters into words and symbols;
assigning a score to each of the words;
calculating a total of the scores assigned to each of the words in the idea;
comparing the total of the scores to a threshold, and if the threshold is met or exceeded, then
identifying a part of speech for each word in the electronic message;
identifying specific dependencies between the words in the electronic message;
determining a sentiment of the electronic message based on a combination of the parts of speech and the specific dependencies between words, where higher scores are assigned to words and groups of words where words having a specific part of speech depend from other words having a specific part of speech;
storing the sentiment in a database residing on a memory device;
associating the sentiment with a personality profile;
identifying the creation of an outgoing message within the private network addressed to a person associated with the personality profile;
generating suggestions for the outgoing message; and
displaying one or more of the suggestions on a display associated with the creation of the outgoing message.

### Claim 2 (depends on 1)

The method of claim 1, wherein scores are totaled across two or more ideas.

### Claim 3 (depends on 1)

The method of claim 1, wherein the scores are functions of subjective feedback from a plurality of persons that review and analyze a subset of all incoming messages.

### Claim 4 (depends on 1)

The method of claim 1, further comprising assigning greater weight to scores assigned to pairs of words, where one of two words is a verb and the other is a noun.

### Claim 5 (depends on 1)

The method of claim 1, wherein the score assigned to a word is greater when the word depends from another word in the idea, where both words match words found in a phrase template.

### Claim 6 (depends on 1)

The method of claim 1, further comprising assigning a score to a group of words in an idea that exceeds a threshold where the group of words comprises at least one verb and at least one object that match a phrase template comprising an array of verbs and objects, where an order of the at least one verb and object and a distance between the at least one object in the digital message and the at least one verb in the digital message are ignored.

### Claim 7

A method of generating computer responses to external communications, the method implemented on a processing device, the method comprising:
recording keystrokes on a first computing device, the keystrokes being part of a digital message;
identifying an end of a first sequence of keystrokes, the first sequence representing an idea;
parsing the keystrokes into words and symbols;
assigning a score to each of the words;
calculating a first total of the scores assigned to each of the words in the idea;
calculating a second total of scores assigned to each word in a plurality of ideas in the digital message, the second total including the first total;
calculating a second total of scores based on parts of speech and dependencies, and where higher scores are assigned where a dependency is between a verb and a direct object;
determining that the second total exceeds a threshold;
precluding the digital message from leaving the first computing device until the digital message is modified to accord with suggested text that is displayed on a display of the first computing device; and then
allowing the digital message to leave the first computing device.

### Claim 8 (depends on 7)

The method of claim 7, wherein the digital message is selected from the group consisting of: a real-time text transmission message, a text message, an electronic mail message.

### Claim 9 (depends on 7)

The method of claim 7, wherein scores are totaled across two or more ideas.

### Claim 10 (depends on 7)

The method of claim 7, wherein the scores are functions of subjective feedback from a plurality of persons that review and analyze a subset of all incoming messages.

### Claim 11 (depends on 7)

The method of claim 7, further comprising assigning greater weight to scores assigned to pairs of words, where one of two words is a verb and the other is a noun.

### Claim 12 (depends on 7)

The method of claim 7, wherein the score assigned to a word is greater when the word depends from another word in the idea, where both words match words found in a phrase template.

### Claim 13 (depends on 7)

The method of claim 7, further comprising assigning a score to a group of words in an idea that exceeds a threshold where the group of words comprises at least one verb and at least one object that match a phrase template comprising an array of verbs and objects, where an order of the at least one verb and object and a distance between the at least one object in the digital message and the at least one verb in the digital message are ignored.

### Claim 14

A non-transitory, tangible computer readable storage medium, encoded with processor readable instructions to perform a method for generating computer responses to external communications, the method comprising:
identifying one or more sequences of characters in an incoming electronic message entering a private network, the sequence representing an idea;
parsing the characters into words and symbols;
assigning a score to each of the words;
identifying certain of the words as belonging to specific parts of speech;
calculating a total of the scores assigned to each of the words in the idea where two words having a dependency between a verb and a direct object are assigned higher scores;
determining a sentiment of the electronic message based on the scores;
storing the sentiment in a database residing on a memory device;
associating the sentiment with a personality profile;
generating an outgoing message in response to the incoming message that is automatically tailored to the sentiment.

### Claim 15 (depends on 14)

The non-transitory, tangible computer readable storage medium of claim 14, wherein the outgoing message is selected from a database of outgoing messages, each associated with one of a list of sentiments.

### Claim 16 (depends on 14)

The non-transitory, tangible computer readable storage medium of claim 14, wherein the digital message is selected from the group consisting of: a real-time text transmission message, a text message, an electronic mail message.

### Claim 17 (depends on 14)

The non-transitory, tangible computer readable storage medium of claim 14, wherein scores are totaled across two or more ideas.

### Claim 18 (depends on 14)

The non-transitory, tangible computer readable storage medium of claim 14, wherein the scores are functions of subjective feedback from a plurality of persons that review and analyze a subset of all incoming messages.

### Claim 19 (depends on 14)

The non-transitory, tangible computer readable storage medium of claim 14, further comprising assigning greater weight to scores assigned to pairs of words, where one of two words is a verb and the other is a noun.

### Claim 20 (depends on 14)

The non-transitory, tangible computer readable storage medium of claim 14, wherein the score assigned to a word is greater when the word depends from another word in the idea, where both words match words found in a phrase template.