# US 12,112,137 B2 · Generating targeted message distribution lists

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

- Number: US 12,112,137 B2
- Document ID: US12112137B2
- Status: Granted
- Kind: B2
- Application number: 17/363,120
- Filed: 2021-06-30
- Published: 2023-01-05
- Granted: 2024-10-08
- Earliest claimed: 2021-06-30
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Jason L. Peipelman; Erica Banda; Rolf Kocheisen; Eric Vermeer; Debora Yang
- Web page: https://patentgrep.com/patent/US12112137B2

## CPC

- G06F 40/30: Semantic analysis
- G06F 40/35: Discourse or dialogue representation
- G06F 40/268: Morphological analysis
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- H04L 51/216: Handling conversation history, e.g. grouping of messages in sessions or threads
- H04L 51/42: Mailbox-related aspects, e.g. synchronisation of mailboxes
- H04L 51/48: Message addressing, e.g. address format or anonymous messages, aliases
- 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

An approach for generating a distribution list. The approach analyzes a message associated with a user to create a message bag of words (BOW). The approach can analyze a history of messages associated with the user to create a plurality of history message BOWs. The approach can calculate similarity factors between the message BOW and the plurality of history BOWs, respectively. If a similarity factor is “>=” a predetermined similarity threshold, the approach can add a history message to a similar message list. The approach can calculate interest factors, based on the similarity factors, for contacts associated with the similarity factors in the similar message list. If an interest factor is “>=” a predetermined interest threshold, the approach can add a contact associated with the interest factor to a suggested contact list. The approach can prioritize the suggested contact list and insert the list as a distribution list.

## Claims

### Claim 1

A computer-implemented method for generating a distribution list, the computer-implemented method comprising:
analyzing a message associated with a user to create a message bag of words (BOW);
analyzing history messages associated with the user to create a plurality of history BOWs;
calculating similarity factors between the message BOW and the plurality of history BOWs, respectively;
persisting the message BOW to reuse and update based on additional messages arriving on a message thread from a contact associated with the message BOW;
responsive to a similarity factor equal to or greater than a predetermined similarity threshold, adding a history message associated with the similarity factor to a similar message list;
calculating interest factors, based on the similarity factors, for contacts associated with the similarity factors in the similar message list, wherein calculating the interest factors further comprises:
scaling an interest factor inversely proportional to a number of the history messages associated with the contact when scaling up is enabled, and
scaling an interest factor proportional to the number of the history messages associated with the contact when scaling down is enabled;
responsive to an interest factor equal to or greater than a predetermined interest threshold, adding a contact associated with the interest factor to a suggested contact list;
prioritizing the suggested contact list based on the interest factor and the number of similarity factor entries in the similar message list of a contact; and
inserting the suggested contact list as a distribution list into a delivery address of an email.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the message BOW is created based on a natural language processing of the message.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the similarity factor is calculated as a weighted average based on a number of similar words between the message BOW and a history BOW divided by the number of words in the history BOW.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the interest factor is calculated as a weighted average based on a sum of a number of history messages associated with a contact in the similar message list divided by a total number of messages in the similar message list.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, further comprising:
adding the contacts from an inclusion list to the suggested contact list; and
removing any contact in the suggested contacts list matching a contact in an exclusion list.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, further comprising:
generating one or more tags associated with the history messages based on the message BOW; and
generating a static distribution list for the message.

### Claim 7

A computer program product for generating a distribution list, the computer program product comprising:
one or more non-transitory computer readable storage media and program instructions stored on the one or more non-transitory computer readable storage media, the program instructions comprising:
program instructions to analyze a message associated with a user to create a message bag of words (BOW);
program instructions to analyze history messages associated with the user to create a plurality of history BOWs;
program instructions to calculate similarity factors between the message BOW and the plurality of history BOWs, respectively;
program instructions to persist the message BOW to reuse and update based on additional messages arriving on a message thread from a contact associated with the message BOW;
responsive to a similarity factor equal to or greater than a predetermined similarity threshold, program instructions to add a history message associated with the similarity factor to a similar message list;
program instructions to calculate interest factors, based on the similarity factors, for contacts associated with the similarity factors in the similar message list, wherein the program instructions to calculate the interest factors further comprise instructions to:
scale an interest factor inversely proportional to a number of the history messages associated with the contact when scaling up is enabled, and
scale an interest factor proportional to the number of the history messages associated with the contact when scaling down is enabled;
responsive to an interest factor equal to or greater than a predetermined interest threshold, program instructions to add a contact associated with the interest factor to a suggested contact list;
program instructions to prioritize the suggested contact list based on the interest factor and the number of similarity factor entries in the similar message list of a contact; and
program instructions to insert the suggested contact list as a distribution list into a delivery address of an email.

### Claim 8 (depends on 7)

The computer program product of claim 7, wherein the message BOW is created based on a natural language processing of the message.

### Claim 9 (depends on 7)

The computer program product of claim 7, wherein the similarity factor is calculated as a weighted average based on a number of similar words between the message BOW and a history BOW divided by a number of words in the history BOW.

### Claim 10 (depends on 7)

The computer program product of claim 7, wherein the interest factor is calculated as a weighted average based on a sum of a number of history messages associated with a contact in the similar message list divided by a total number of messages in the similar message list.

### Claim 11 (depends on 7)

The computer program product of claim 7, further comprising:
program instructions to add the contacts from an inclusion list to the suggested contact list; and
program instructions to remove any contact in the suggested contacts list matching a contact in an exclusion list.

### Claim 12 (depends on 7)

The computer program product of claim 7, further comprising:
program instructions to generate one or more tags associated with the history messages based on the message BOW; and
program instructions to generate a static distribution list for the message.

### Claim 13

A computer system for generating a distribution list, the computer system comprising:
one or more computer processors;
one or more non-transitory computer readable storage media; and
program instructions stored on the one or more non-transitory computer readable storage media, the program instructions comprising:
program instructions to analyze a message associated with a user to create a message bag of words (BOW);
program instructions to analyze history messages associated with the user to create a plurality of history BOWs;
program instructions to calculate similarity factors between the message BOW and the plurality of history BOWs, respectively;
program instructions to persist the message BOW to reuse and update based on additional messages arriving on a message thread from a contact associated with the message BOW;
responsive to a similarity factor equal to or greater than a predetermined similarity threshold, program instructions to add a history message associated with the similarity factor to a similar message list;
program instructions to calculate interest factors, based on the similarity factors, for contacts associated with the similarity factors in the similar message list, wherein the program instructions to calculate the interest factors further comprise instructions to:
scale an interest factor inversely proportional to a number of the history messages associated with the contact when scaling up is enabled, and
scale an interest factor proportional to the number of the history messages associated with the contact when scaling down is enabled;
responsive to an interest factor equal to or greater than a predetermined interest threshold, program instructions to add a contact associated with the interest factor to a suggested contact list;
program instructions to prioritize the suggested contact list based on the interest factor and the number of similarity factor entries in the similar message list of a contact; and
program instructions to insert the suggested contact list as a distribution list into a delivery address of an email.

### Claim 14 (depends on 13)

The computer system of claim 13, wherein the message BOW is created based on a natural language processing of the message.

### Claim 15 (depends on 13)

The computer system of claim 13, wherein the similarity factor is calculated as a weighted average based on a number of similar words between the message BOW and a history BOW divided by a number of words in the history BOW.

### Claim 16 (depends on 13)

The computer system of claim 13, wherein the interest factor is calculated as a weighted average based on a sum of the number of history messages associated with a contact in the similar message list divided by a total number of messages in the similar message list.

### Claim 17 (depends on 13)

The computer system of claim 13, further comprising:
program instructions to add the contacts from an inclusion list to the suggested contact list; and
program instructions to remove any contact in the suggested contacts list matching a contact in the exclusion list.