# US 2023/0274745 A1 · Natural language based electronic communication profile system

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

- Number: US 2023/0274745 A1
- Document ID: US20230274745A1
- Status: Published application
- Kind: A1
- Application number: 18/312,155
- Filed: 2023-05-04
- Published: 2023-08-31
- Earliest claimed: 2020-06-08
- Assignees: Capital One Services, LLC
- Applicants: Capital One Services, LLC
- Inventors: Christopher CAMENARES; Joseph BOAYUE; Lee ADCOCK; Ana CRUZ; Nahid Farhady GHALATY
- Web page: https://patentgrep.com/patent/US20230274745A1

## CPC

- G10L 15/26: Speech to text systems (G10L15/08 takes precedence)
- G06F 40/289: Phrasal analysis, e.g. finite state techniques or chunking
- G06F 40/30: Semantic analysis
- G10L 15/1822: {Parsing for meaning understanding}
- G06F 21/30: Authentication, i.e. establishing the identity or authorisation of security principals
- G06F 21/316: {by observing the pattern of computer usage, e.g. typical user behaviour}
- G06F 40/211: Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
- G06F 40/253: Grammatical analysis; Style critique
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- G10L 17/02: Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Systems, methods, and computer-readable storage media for generating a quantifiable communication profile of a participant, then using that communication profile to identify fake media and/or verify media as authentic. An exemplary system can receive communications in which a participant was one of the parties communicating, perform natural language processing and statistical analyses on the received communications, and generate a participant communication profile which details how the participant communicates and prefers to be communicated with. The system can also receive media purported to be from the participant, perform natural language processing on the purported media, compare the syntax of the purported media to the participant’s communication profile, and identify, based on word usage, the media as a fake.

## Claims

### Claim 1

A method, comprising:
receiving, from a database of stored communications, a plurality of communications in which a user was one of the parties communicating;
performing, via a processor on each communication within the plurality of communications, natural language processing, the natural language processing resulting in communication data generated by the user;
executing, via the processor, a statistical analysis on the communication data, the statistical analysis comprising at least one regression, resulting in statistical communication data of the user;
generating, via the processor and using the statistical communication data of the user, a communication profile of the user, the communication profile comprising a data structure storing a user identifier associated with a personalized linguistic syntax of the user and the statistical communication data of the user;
receiving media containing audio purported to be from the user;
performing, via the processor, natural language processing on the audio purported to be from the user, resulting in analyzed purported audio;
comparing, via the processor, a syntax of the analyzed purported audio to the personalized linguistic syntax stored within the communication profile of the user, resulting in a syntax similarity score;
comparing, via the processor, words used within the analyzed purported audio with the statistical communication data of the user, resulting in a word usage similarity score; and
identifying, via the processor and based on the syntax similarity score and the word usage similarity score, the media as a fake.

### Claim 2 (depends on 1)

The method of claim 1, the communication data comprising, for each communication: a topic;
identification of any slang used by the user;
a sentence structure used by the user; and
an identity of at least one other person communicating with the user.

### Claim 3 (depends on 1)

The method of claim 1, wherein the plurality of communications comprise tagged electronic transcripts, and wherein the natural language processing is executed on the tagged electronic transcripts.

### Claim 4 (depends on 3)

The method of claim 3, wherein the tagged electronic transcripts contain time stamps.

### Claim 5 (depends on 1)

The method of claim 1, wherein the plurality of communications comprise audio recordings, and
wherein the natural language processing executes a speech-to-text conversion in generating the communication data.

### Claim 6 (depends on 1)

The method of claim 1, wherein the statistical communication data of the user comprises:
average words per minute by the user for each communication in the plurality of communications;
average duration of each communication in the plurality of communications;
at least one of graphics or emojis used by the user.

### Claim 7 (depends on 6)

The method of claim 6, wherein:
the performing of the natural language processing further comprises:
identifying, via the processor for each communication in the plurality of communications, a communicator of the communication and at least one recipients of the recipient based on at least one of a voice profile of individuals participating in the communication or previously generated communication profiles of individuals participating in the communication, wherein the user is either the communicator or a recipient within the at least one recipients; and
wherein execution of the statistical analysis further comprises:
calculating, via the processor, the average words for a unit of time of the user for both when the user is identified as a communicator and when the user is identified as a recipient.

### Claim 8 (depends on 1)

The method of claim 1, further comprising:
transmitting, via the processor, the communication profile to a user device of the user;
receiving, at the processor from the user device, feedback associated with the communication profile; and
modifying, via the processor, code executed by the processor regarding the generating of the communication profile based on the feedback.

### Claim 9 (depends on 1)

The method of claim 1, wherein the communication profile identifies quantitative behavior of the user within the plurality of communications.

### Claim 10 (depends on 9)

The method of claim 9, wherein the communication profile is iteratively modified as new communications in which the user is a participant are added to the database of stored communications.

### Claim 11

A system for quantifying user communication behavior, the system comprising:
an input device;
a display device;
a processor;
a non-transitory computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
receiving, from the input device, a communication from a user, the communication comprising one of an email, an audio clip, or a SMS (Short Message Service) message;
performing natural language processing on the communication, resulting in a parsed communication having a data structure comprising: (1) the communication, and (2) identification of at least one of nouns, verbs, pronouns, emoticons, or phrases within the communication;
identifying, using the at least one of nouns, pronouns, emoticons, or phrases stored within the data structure of the parsed communication, a context for the communication, wherein the context is identified by the processor comparing the at least one of nouns, pronouns, emoticons, or phrases stored in the data structure to at least one of nouns, pronouns, emoticons, or phrases associated with predetermined contexts and, upon identifying a similar usage of the parsed communication and the predetermined contexts, identifying the context based on the similar usage;
tagging the parsed communication with a contextual tag based on the context, resulting in a tagged parsed communication;
transmitting the tagged parsed communication to a database;
receiving, from a server which is in electronic communication with the database, a communication profile of the user, wherein the communication profile is based at least in part on the tagged parsed communication; and
displaying, via the display device, the communication profile.

### Claim 12 (depends on 11)

The system of claim 11, the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
updating, within a database, the predetermined contexts by adding the tagged parsed communication to the predetermined contexts, resulting in updated predetermined contexts, such that future determinations of contexts use the updated predetermined contexts.

### Claim 13 (depends on 11)

The system of claim 11, the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
receiving, from the user via the input device, a privacy setting for the communication profile, the privacy setting identifying specific statistics within the communication profile which are exclusively visible to the user.

### Claim 14 (depends on 13)

The system of claim 13, the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
transmitting the privacy setting to the server, such that other parties communicating with the user may access a redacted version of the communication profile of the user.

### Claim 15 (depends on 11)

The system of claim 11, wherein the communication profile comprises:
an average words per minute by the user;
identification of slang used by the user;
identification of preferred topics of the user; and
identification of at least one of graphics or emojis used by the user.

### Claim 16 (depends on 11)

The system of claim 11, wherein the communication profile is generated by the server based on the tagged parsed communication and previous tagged parsed communications of the user.

### Claim 17 (depends on 11)

The system of claim 11, the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
receiving, via the input device, feedback associated with the communication profile; and
transmitting, from the processor to the server, feedback regarding the communication profile.

### Claim 18 (depends on 17)

The system of claim 17, the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
receiving, from the server, code modifications based on the feedback, the code modifications changing how the processor parses future communications;
executing, via the processor, the code modifications on the non-transitory computer-readable storage medium, resulting in a modified non-transitory computer-readable storage medium; and
processing future communications from the user using code from the modified non-transitory computer-readable storage medium.

### Claim 19 (depends on 11)

The system of claim 11, the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
receiving, from the server, code modifications, the code modifications reducing an amount of natural language processing required for the user based on the communication profile;
executing, via the processor, the code modifications on the non-transitory computer-readable storage medium, resulting in a modified non-transitory computer-readable storage medium; and
processing future communications from the user using code from the modified non-transitory computer-readable storage medium.

### Claim 20

A non-transitory computer-readable storage medium having instructions stored which, when executed by a computer, cause the computer to perform operations comprising:
receiving, from user device, a tagged communication in which a user was communicating;
adding the tagged communication to a database of communications, the database of communications storing a plurality of tagged communications from the user;
performing, on each tagged communication within the plurality of tagged communications, natural language processing, the natural language processing resulting in user communication data;
executing a regression analysis on the user communication data, resulting in weights identifying how the user communicates;
generating, using the weights identifying how the user communicates, a quantitative profile of how the user communicates; and
transmitting the quantitative profile of how the user communicates to the user device.
